EEG Pearls

1600 John F. Kennedy Boulevard Suite 1800 Philadelphia, PA 19103-2899

EEG PEARLS Copyright # 2006 by Mosby, Inc., an affiliate of Elsevier, Inc.

ISBN-13: 978-0-323-04233-8 ISBN-10: 0-323-04233-3

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior permission of the publisher (WB Saunders, 1600 John F. Kennedy Boulevard, Suite 1800, Philadelphia, PA 19103–2899).

NOTICE Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment and drug therapy may become necessary or appropriate. Readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of the practitioner, relying on his or her own experience and knowledge of the patient, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the Editor assumes any liability for any injury and/or damage to persons or property arising out or related to any use of the material contained in this book. The Publisher

Library of Congress Cataloging-in-Publication Data EEG pearls / [edited by] Mark Quigg.–1st ed. p. ; cm. ISBN 0-323-04233-3 1. Electroencephalography–Case studies. 2. Electroencephalography–Problems, exercises, etc. I. Quigg, Mark. [DNLM: 1. Electroencephalography–Case Reports. 2. Electroencephalography–Problems and Exercises. 3. Nervous System Diseases–diagnosis–Case Reports. 4. Nervous System Diseases–diagnosis–Problems and Exercises. WL 18.2 Q6e 2006] RC386.6.E43E13 2006 616.8’047547–dc22 2006041975

Acquisitions Editor: Linda Belfus Developmental Editor: Stan Ward Project Manager: David Saltzberg

Printed in the United States of America. Last digit is the print number: 9

8

7

6 5

4

3 2

1

For Lotta, Anders, and Erik

ix

Acknowledgments The author gratefully acknowledges the expert editorial assistance provided by the following colleagues: HOWARD GOODKIN M.D., PH.D. Assistant Professor of Neurology University of Virginia Charlottesville, Virginia MARK A. GRANNER M.D. Associate Professor of Clinical Neurology Roy A. and Lucille J. Carver College of Medicine University of Iowa Iowa City, Iowa WILLIAM R. HOBBS M.D. Professor of Psychiatry University of Virginia Charlottesville, Virginia JAMES Q. MILLER M.D. Professor of Neurology University of Virginia Charlottesville, Virginia The author also thanks the dedicated EEG technologists at the laboratory of the University of Virginia. Their high-quality work and helpful comments made this work possible.

xi

BASIC PRINCIPLES OF ELECTRICITY 1

Basic Electricity

Engineering advances allowed Hans Berger to record the first human electroencephalogram (EEG) in the late 1920s. Since then, EEG has evolved to become an important tool in the evaluation of epilepsy and encephalopathy. The basic task of the EEG machine is the faithful detection of the electrical activities generated by the brain. Many details of the human EEG can be memorized. However, to understand EEG, to rationalize its behaviors, and to understand its confounders, an elementary knowledge of electricity and EEG technology is needed (Fig. 1-1).

Figure 1-1. Basic science of electricity.

A charge (Q, coulomb) is the basic unit of electricity. One coulomb is equal to the total charge of 6  1018 electrons. Movement of electrons from place to place creates current (I, amperes, or amps). One amp (A) of current represents the flow of one coulomb of electrons during 1 s. The electrical impetus that forces current from place to place is voltage or potential (V, volts). Voltage measures the energy applied to a unit of charge (V ¼ energy/charge). An analogy to water flow is a useful way to conceptualize electrical properties. Current flows ‘‘downhill’’ from regions of high gravitational potential to regions of low potential. Electrical potential is always measured as a comparison between two points. The electrical reference equivalent to atmospheric pressure at sea level is the electrical ground, the theoretical lowest potential within the substance of the earth. The flow of current through a wire is impeded by resistance (O, ohms). The amount of current that can squeeze through a restriction—an electrical resistor—is related to the voltage that can be mustered to force it past the restriction. A small voltage can push a small current, and a large voltage can move a torrent. Similarly, a large resistor will cause a large drop in potential as current forces its way through, whereas a small resistor causes only a small loss. These relationships are represented by Ohm’s law: V¼IR

1

Questions: 1. What is the term for the theoretical point of lowest potential? 2. What is the drop in potential at a 5 kO resistor for a current of 10 pA? Express the answer in units of mV. 3. What is the product of voltage and charge? Answers: 1. The theoretical lowest potential is electrical ground. The measurement of electrical potential is always a comparison of potentials between two points. 2. V ¼ I  R ¼ 5 kO  10 pA ¼ 5  103O  10  109 A ¼ 50  106 V ¼ 50 mV 3. Energy ¼ voltage  charge

Pearls 1. Voltage is always measured as the difference in potential between two points. 2. V ¼ I  R Voltage ¼ current  resistance 3. Energy ¼ voltage  charge

2

Basic Principles of Electricity

Elementary Circuits

When voltage remains constant for long periods of time, the current likewise remains constant. A common flashlight is an example of a direct current (DC) circuit, with a steady voltage supply (the battery) driving a constant current across the steady resistance of the flashlight bulb (Fig. 1-2).

Figure 1-2. DC and AC circuits.

When voltage fluctuates over time, an alternating current (AC) circuit results. Household current in the United States alternates at a frequency of 60 cycles per second (cycles per second ¼ Herz, or Hz). The period (analogous to wavelength) is the time from peak to peak of each cycle and is the reciprocal of frequency: Period ¼ 1=f The phase is the reference point, usually the peak of the cycle, measured in relation to an initial point in time. Household current oscillates rapidly between high and low potentials, in effect, pulling and pushing electrons back and forth. Most biologic signals form AC circuits, with the fluctuations of cations and anions moving across cell membranes playing the role of electrons oscillating within a wire. The fluctuating nature of an AC circuit requires the addition of two circuit properties: capacitance and inductance. Impedance (Z, ohms, O) is the combined effect of capacitance, inductance, and resistance on AC current flow. A capacitor consists of two conducting surfaces separated by a nonconducting insulator, such as a sandwich of two plates of metal separated by a rubber sheet. Inserted into a simple DC circuit, the capacitor allows the buildup of electrons on the plate nearest the voltage source until the mutual repulsion of the collected electrons begins to counterbalance the strength of the voltage source. Therefore, the flow of current gradually stops when the capacitor is ‘‘full.’’ The more charge a capacitor can hold for a given voltage, the greater the capacitance (C, farad), given by the following equation: C ¼ Q=V ¼ charge=voltage The effect of the capacitor is strikingly different when inserted into an AC circuit. As in the DC circuit, current flows until the capacitor is fully charged. However, when the AC power source fluctuates, and the potential pushing the electrons to the capacitor abruptly drops, the stored electrons are free to exit the capacitor in the opposite direction from which they entered. The current reverses direction. For an AC circuit, therefore, as long as the source of voltage fluctuates, a capacitor never Elementary Circuits

3

totally blocks current flow, as it does in a DC circuit, because electrons continue to collect and disperse alternately on each side of the capacitor. The contribution of capacitance to the overall impedance of an AC circuit depends on the frequency of the alternating current. The effective resistance of a capacitor to current flow is capacitive reactance (Xc) and is inversely proportional to the frequency and the capacitance: Xc ¼ 1=ð2p  f  CÞ Capacitive reactance to a current with a frequency of zero (a DC current) is infinite. As frequency increases, the capacitive reactance drops, allowing more current at the higher frequency to be pushed and pulled across the capacitor. Inductance, although important in everyday electrical devices (electric motors are powered by induction of magnetism by fluctuating current), has negligible effect on EEG. Questions: 1. What is the period of a current of 10 mA carried at a frequency of 50 Hz? 2. What constitutes a capacitor? 3. What is the relationship between frequency and capacitive reactance? 4. What is the capacitive reactance of a current with frequency of zero? 5. What is the impedance for a 25-Hz signal that generates 100 mV at a current of 0.02 mA? Express the answer in units of k. Answers:

1. Period ¼ 1/f ¼ 1/50 Hz ¼ 0.02 s ¼ 20 ms 2. A capacitor consists of two conducting surfaces separated by an insulator. In effect, any electrical junction between dissimilar materials can act as a capacitor. Capacitive reactance is important in EEG; one example is the impedance caused by the junction between the EEG electrode and the scalp. Oil, dirt, or dandruff, for example, could act as an insulator between the two conducting surfaces. Differences in impedance among electrodes can affect the quality of the recording. 3. Xc ¼ 1/f. Note that signal frequency is also inversely proportional to the capacitance, a relationship essential in the design of EEG filters. 4. Capacitive reactance of a current with frequency of zero is infinite. 5. Z ¼ V/I ¼ 100 mV/0.02 mA ¼ 5000 O ¼ 5 kO

Pearls 1. Resistance to current flow in an AC circuit is called impedance and is proportional to resistance and capacitive reactance. 2. Capacitive reactance is inversely proportional to frequency. 3. A capacitor is formed at any electrical junction. In the case of EEG, the most important contribution to impedance is the connection between scalp and electrode.

4

Basic Principles of Electricity

Electrical Safety

Any time two electrodes are attached to a subject and both electrodes are connected to a measuring device, the subject becomes a possible pathway for current. The safety of this biologic circuit element should not be taken for granted. Memorization of simple rules will keep everyone out of trouble. First, electrical medical instruments must adhere to electrical safety requirements and must be inspected and approved by clinical engineering departments before use. Second, medical instruments are grounded to earth. A defect in wiring, such as a frayed wire touching the metal instrument cabinet, can allow current to leak. Because current follows the path of least resistance, a good ground allows the current to flow away from, rather than through, the subject, who in contrast, offers a much higher resistance to current flow. Third, the subject should not be exposed to earth ground. All electrical instruments that attach to the patient require a ground, from electrocardiogram (EKG) monitors to EEG machines to electrocautery devices. These instrument-patient connections are isolated grounds; in other words, although the subject and instrument achieve the same overall ground potential, the subject is not tied, in turn, to the main earth ground. Tying the subject to earth ground can be dangerous; the patient, in this case, becomes part of a low impedance circuit that can carry inadvertent current through the patient. Indeed, modern EEG systems totally isolate the patient from external current sources through a low capacitance barrier, through optical-electrical transducers, or other engineering means. Fourth, the subject should not be exposed to multiple grounds. Although ground denotes the lowest possible electrical potential, the ground potential at one location may not exactly coincide with another at all times. When our sea level analogy to ground potential is applied, small waves come and go that minutely change the level of water. The presence of two grounds, either through differences in impedance in their connection to the body or by fluctuations in ground potentials among different sources, can allow current to flow from ground to ground. Different instruments, therefore, connect to a common patient ground. Another advantage of a common ground is that, often, electrical noise is minimized, enabling a clean recording. Question: What is the artifact of the tracing below (Fig. 1-3) at electrode F3? Note that because channels Fp1-F3 and F3-C3 share a common faulty electrode, the noise it generates will appear in both.

Figure 1-3. Artifact at electrode F3.

Electrical Safety

5

Answer:

60 Hz electrical artifact.

Pearls 1. All medical devices must be approved by an appropriate clinical engineering department. 2. Avoid multiple grounds, both to avoid noise marring the recording and to minimize the possibility of electrocution.

6

Basic Principles of Electricity

Filters

EEG machines use filters to dampen extraneous potentials. Before computerization and the advent of the digital EEG, filters were constructed from combinations of capacitors and resistors (Fig. 1-4).

Figure 1-4. Low-frequency and high-frequency filters.

The key to understanding filters is to recall the relationship between signal frequency and capacitive reactance. It takes time for a capacitor to store up or discharge current, a duration measured by the time constant (t in seconds). For any given resistance R and capacitance C, t ¼ R  C. Capacitors charge or discharge at an exponential rate. Whereas biologists use half-life to describe exponential relationships, engineers use the natural logarithm e. The time constant for an RC circuit is the duration required for the output signal to discharge to 37% (1/e) of the input signal. In succinct mathematical terms, for an RC circuit in which Vo ¼ output voltage and Vi ¼ input voltage: Vo ¼ Vi  et=t A low-frequency filter (LFF, sometimes called a high-pass filter) uses a capacitor wired in series with a resistor. High-frequency signals can pass through the capacitor because the capacitive reactance is small for rapidly alternating currents. Low frequencies, on the other hand, are more easily blocked because capacitive reactance rises with decreasing frequency. Low-frequency filters remove low frequency artifact, such as potentials generated from slight temperature changes or skin conductance (galvanic potentials), tissue-electrode polarization, and patient movement. A high-frequency filter (HFF, or low-pass filter) uses a capacitor wired in parallel with a resistor. High-frequency signals preferentially shunt through the capacitor because the capacitive reactance is small in comparison to resistance. Conversely, low-frequency signals ‘‘see’’ a large capacitive reactance and proceed along the easier path through the resistor. High-frequency filters remove artifact, such as muscle noise from the signal. A notch filter uses combinations of RC circuits to remove specific frequencies from a signal. The typical application for a notch filter is the removal of 60-Hz electrical noise from the signal. The signals for display in digital EEG also undergo filtering, but RC circuits are replaced by mathematical functions that manipulate frequency spectra. Nevertheless, most manufacturers maintain terminology and function derived from traditional analog recording methods. Questions: 1. What is the relationship between resistance, capacitance, and time constant in an RC circuit? 2. The time constant measures the time it takes for a capacitor to discharge by what percentage of its initial value? Filters

7

3. Does an HFF remove high frequencies by putting a capacitor in series or in parallel with a resistor? Answers: 1. t ¼ R  C 2. 37% of discharge. Note that t also designates the time a capacitor charges to 63% of the maximum voltage. 3. Parallel.

Pearls 1. The time constant measures how long an RC circuit takes to charge and discharge a capacitor, with longer time constants implying larger capacitors that offer a lower capacitive reactance. Thus, the shorter the time constant, the more difficulty lowfrequency signals will have traversing an RC circuit. 2. Low-frequency filters have a capacitor in series with a resistor and remove lowfrequency signals. The standard setting for LFF is 1 Hz. 3. High-frequency filters have a capacitor in parallel with a resistor and remove high-frequency signals. The standard settings for HFF are 70 Hz or 35 Hz (varies with manufacturer).

8

Basic Principles of Electricity

Filters and Cutoff Frequency

The amount of filtering applied to an EEG signal is specified by the filter’s cutoff frequency ( fcutoff) or time constant (t). Traditionally, time constant and cutoff frequency are interchangeably used to designate the LFF setting, whereas the cutoff frequency designates the HFF setting. The cutoff frequency can be calculated from the time constant by the following equation: fcutoff ¼ ðp=2Þ  1=t In the tracing in Figure 1-5, different filter settings are applied to the same signal. Manufacturers designate the cutoff frequency settings on their EEG machines by noting the setting at which 70% of the signal at the cutoff frequency passes through the filter. For example, an LFF setting of 1 Hz denotes that frequencies ¼ 1 Hz will be attenuated by at least 30%, and frequencies 30%. Such relationships are easily summarized by a frequency-response graph (Fig. 1-6) that plots the rate of attenuation of output signal at different filter settings.

Figure 1–6. Frequency-response curve.

Pearls 1. Time constant is reciprocally related to the filter cutoff frequency. 2. The cutoff frequency indicates the frequency above or below which 70% of the input voltage is allowed to pass on to display. 3. Routine starting cutoff frequencies are LFF ¼ 1 Hz and HFF ¼ 70 Hz or 35 Hz.

10

Basic Principles of Electricity

Sensitivity and Paper Speed

Sensitivity defines the amplitude of the EEG display signal: how large a given EEG potential displays upon the paper or computer screen (Fig. 1-7). Analog and digital EEG systems differ in the units of sensitivity.

Figure 1-7. Sensitivity defines the amplitude of the EEG display signal.

The units of sensitivity of an analog EEG system are given in mV/mm, thus enabling the interpreter to easily calculate the amplitude of a potential by measuring its height in millimeters and multiplying by the sensitivity value. For example, Potential (mVÞ ¼ height (mmÞ  sensitivity (mV=mmÞ ¼ 10 mm  7 mV=mm ¼ 70 mV Digital EEG systems divide the display into the number of vertical pixels allowed for each channel. The units of sensitivity of a digital EEG system are given in peak-to-peak (p-p) microvolts per channel. A sensitivity of 150 mV p-p, therefore, specifies that the maximum potential fully visible in that channel is 150 mV. The trade-off for the ease in measurement is that the number designating sensitivity is reciprocal to its effect; in other words, the same signal displayed at a sensitivity of 10 mV/mm (or its approximate digital equivalent of 300 mV p-p) will be displayed smaller than it would be at a sensitivity of 2 mV/mm (or around 50 mV p-p). Sensitivity, analogous to the volume level of a stereo, has no inherently correct value and is set best to display potentials at the most informative level. Usually, the level appropriate for most adult studies is 7 mV/mm (or 150 mV p-p). Sensitivities that are too high cause blocking (so-called because the sweep of the EEG pens allows them to hit one another) or clipping (in the case of digital EEG) and are to be avoided in recording of high-amplitude potentials. The EEG technologist must label the tracing whenever changing sensitivity so that the interpreter makes no mistakes in comparing the amplitudes of potentials across different sections. For scalp recordings, the maximum sensitivity is 2 mV/mm; signals with amplitude below 1 mm at this setting are considered noise. Standard paper speed of EEG recordings in the United States is 30 mm/s (Fig. 1-8). Faster speeds of 60 mm/s are sometimes used intermittently during a tracing to closely examine high-frequency activity or closely spaced potentials. Slower paper speeds of 15 mm/s allow conservation of paper and Sensitivity and Paper Speed

11

Figure 1-8. Paper speeds of 10 mm/s and 30 mm/s (standard speed in the United States).

facilitate the study of slower frequencies. Digital EEG systems allow changing of these parameters (as well as others) on the fly. Most EEG systems, paper or digital, display EEG as 10-second pages, with major divisions denoting 1 s and 5 minor subdivisions of 200 ms each. Examples in this book omit minor divisions for clarity. Questions: 1. What is the amplitude in mV of a signal measured on a paper EEG with height ¼ 7 mm at a sensitivity of 7 mV/mm? 2. What is the maximum sensitivity of a standard tracing? Answers:

1. Amplitude ¼ measurement  sensitivity ¼ 7 mm  7 mV/mm ¼ 49 mV Most calibration signals for analog EEGs are 50 mV; therefore, the calibration signal measures around 7 mm at the standard sensitivity of 7 mV/mm. 2. The traditional maximum limit of sensitivity is 2 mV/mm. At this sensitivity on scalp recordings, signals below 1 mm in amplitude are considered noise.

Pearls 1. Sensitivity determines the display size of signals on paper (analog mV/mm) or on the display screen (mV p-p). 2. The standard sensitivity for analog tracings is 7 mV/mm, and the maximum is 2 mV/mm. 3. Standard paper speed is 30 mm/s displayed in 10-second pages. 4. Technologists must annotate the tracing so that changes in sensitivity, filter settings, paper speed, or montages are clearly observable by the interpreter. REFERENCE 1. American Electroencephalographic Society. Guideline one: Minimum technical requirements for performing clinical electroencephalography. J Clin Neurophysiol 1994; 11:2–5.

12

Basic Principles of Electricity

Signal Processing

Digital EEG is rapidly becoming the standard in American clinical neurophysiology laboratories. The advantages in storage, data analysis, and display over traditional analog systems will become evident with the examples in this book. An understanding of analog-to-digital (ATD) conversion is necessary to understand limitations of this technique. Analog data, the traditional voltage-time data displayed by the EEG, are smoothly and continuously variable. Digital data, on the other hand, are converted into discrete, stepped values. The finer the steps, the more accurately digital data represent analog data. Two variables, sampling frequency and bit depth, determine the accuracy of ATD conversion (Fig. 1-9).

Figure 1-9. Analog and digital signal processing.

Sampling frequency determines the number and density of samples taken along the time axis. A sampling frequency of 100 Hz, therefore, reads 100 values for voltage data consecutively for each second of data. Sampling frequency determines the size of data files: a sampling frequency of 200 Hz is more accurate than 100 Hz but creates data files twice the size. The accuracy of sampling is limited by Nyquist’s theorem: Sampling frequency ¼ 2  fmax In other words, the fastest frequency signal that can accurately be represented is one-half the sampling frequency. For example, a sampling frequency of 256 Hz accurately represents signal frequencies up to 128 Hz. Aliasing occurs when high-frequency signals are misrepresented as slower-frequency signals because they are undersampled. Bit depth, sometimes referred to as vertical resolution, designates the number of divisions into which the range of voltages can be represented. Because computer memory is binary-based, each stepwise increase in bit depth increases the number of possible amplitude levels by a factor of 2. For example, a bit depth of 8 means that there are 28 ¼ 256 individual steps between the minimum and maximum allowable voltages. A bit depth of 9 provides an amplitude resolution of 29 ¼ 512. Question: 1. Given that the clinically relevant range of EEG frequencies is about 0.5–30 Hz, what is the minimum appropriate sampling frequency for routine digital EEG?

Signal Processing

13

Answer: 1. Trick question. By Nyquist’s theorem, to accurately represent signals of 30 Hz, the minimum sampling frequency is 60 Hz. However, the range of clinically relevant EEG frequencies is a different question from the range necessary to represent the morphology of EEG signals accurately. A sampling frequency of 60 Hz limits the shortest duration that can be accurately measured to the reciprocal of 60 ¼ 1/60 ¼ 0.01667 s  17 ms. Because some interictal epileptiform discharges have durations of approximately 20 ms, a sampling frequency of 60 Hz could completely miss some discharges or render others as jagged steps in the tracing. Digital EEG systems use sampling frequencies between 100 and 500 Hz. A typical bit depth is usually 16 (216 ¼ 65536, or in the common computer convention, 64k).

Pearls 1. The highest accurately portrayed frequency in a digital EEG is one-half the sampling rate. 2. Increasing bit depth or vertical resolution of digital data by one step requires a doubling of storage media.

14

Basic Principles of Electricity

The Differential Amplifier

An amplifier is a device that receives an input signal, guides a power source, and creates an amplified copy of the original signal. The increase in voltage is called gain and is calculated from the ratio of the output and input voltages. Often gain is expressed as the engineering term decibels (db): db ¼ 20  log (Vout=VinÞ Thus, a gain of one magnitude (Vout/Vin ¼ 10) is 20 db, two magnitudes (Vout/Vin ¼ 100) ¼ 40 db. Most amplifiers used in recordings of biologic signals are a combination of amplifiers called a differential amplifier. The output of a differential amplifier is the amplified difference between two inputs, called G1 and G2. (‘‘G’’ comes from the days when the electrical contacts to tube amplifiers were grids) (Fig. 1-10). The main benefit of the differential amplifier is noise reduction, quantified by the term common mode rejection. Any voltage seen in common between G1 and G2 adds up to zero and thereby cancels out. One can calculate the common mode rejection ratio (CMRR) by shorting G2 to ground and taking the ratio of the input and output signals. The higher the CMRR, typically 105, the better the quality of the differential amplifier.

Figure 1-10. The output of a differential amplifier is the amplified difference between two inputs, called G1 and G2.

Common mode rejection works most effectively if the impedances of inputs at G1 and G2 are equally matched. In the example, noise, represented as the sinusoidal input signal, is recorded with equal voltages at G1 and G2 because the impedances of G1 and G2 are similar. In this case, noise cancels out. In the case of impedance mismatching, the ‘‘bad electrode’’ at G2 records noise at a higher voltage than G1. Because G1 and G2 transmit unequal voltages, the noise no longer cancels out. The recommended maximum impedance of scalp electrodes in EEG is 5kO. Questions: 1. What is the amplification factor for an amplifier with a gain of 120 db? 2. What is the maximum scalp electrode impedance in EEG? 3. What are the designators for inputs to a differential amplifier?

The Differential Amplifier

15

Answers:

1. Gain: db ¼ 20  log (Vout/Vin) 120 db ¼ 20  log (Vout/Vin) 6 ¼ log (Vout/Vin) 106 ¼ Vout/Vin 2. Maximum scalp electrode impedance ¼ 5 kO 3. G1 and G2 are the designators of the inputs to a differential amplifier.

Pearls 1. Gain is the amplification factor of differential amplifiers, often measured in db. 2. Common mode rejection is the noise-reduction design of the differential amplifier. The higher the common mode rejection ratio, the better. 3. G1 and G2 are the names given by convention for each input pair of electrodes of the differential amplifier. 4. Because impedance mismatch can cause amplification of degraded signal, the maximum impedance of scalp electrodes in EEG is 5kO.

16

Basic Principles of Electricity

ACQUISITION OF THE ELECTROENCEPHALOGRAM 2

Electroencephalographic Electrodes

The collection of data from the patient to the electroencephalographic (EEG) machine starts at the EEG electrode. Soo Ik Lee (MD, former director of the EEG laboratory at the University of Virginia) believed that bad electrodes are like bad teenagers. They either distort the truth or completely make up stories when it suits them. If left uncorrected, they get worse. Similarly, bad electrodes distort brain electrical activity and create their own signals if mismanaged, and only get worse if untended. The goal of electrode placement and maintenance is to make the conduction of electrical current from scalp to machine as accurate as possible. The process starts at the scalp with a scrubbing of the electrode site with a pumice-laced detergent that lightly abrades the skin and removes oil. A conductive gel containing salts in a viscous medium is applied so that ions can carry current between the electrode and skin. Silver-chloride electrodes, named so because the silver has been purposely oxidized with a chloride solution, facilitate conversion of ionic current flow to electron current flow. Silver and chloride ions on the electrode surface are free to pass into the gel solution. However, an oil layer left on the scalp that separates skin and gel creates the equivalent of a large capacitor (two conductors separated by an insulator). A capacitor in the current path raises impedance of the electrode, artificially increasing its signal relative to its neighbors (impedance mismatch). The capacitor also acts as a low-frequency filter, further distorting signal. Scalp electrodes should have a maximum impedance of G2 (G2 more negative), then there is a downward deflection in the channel. 3. If G1 < G2 (G1 more negative), then there is an upward deflection in the channel. 4. Three montages (longitudinal and transverse bipolars, referential) are required for every standard EEG. REFERENCE 1. American Electroencephalographic Society: Guideline one: Minimum technical requirements for performing clinical electroencephalography. J Clin Neurophysiol Electroencephalographic Society 1994; 11:2–5.

Channels and Montages

23

Localization 1

Localization is the process of identifying the polarity and location of an EEG finding. Localization is possible with scalp EEG because of an important simplification. For all of the complex structures of gyri, sulci, and fissures of the brain, its electrical properties can be represented as a smooth hemisphere. This model has several important implications. First, although all electrical potentials form dipoles with one end positive and the other negative, often scalp electrodes only record from one half of the dipole; the other half projects to the lower hemisphere and is not recordable. Second, because the brain has a complex structure, potentials that occur within certain regions may occur without showing on the scalp. For example, the base of the frontal lobe, because it is relatively far away from scalp electrodes, may harbor occult potentials. In the examples below, the electrical dipole is represented as a þ or –. Isopotential lines drawn on the scalp designate regions with equal polarity that drop in intensity with distance from the source. Such maps define an electrical field, the distribution of influence that an individual discharge imparts upon a conductive volume. The EEG for each potential is shown in two montages, a longitudinal bipolar and a referential to an ear. In Figure 2-7, a negative potential occurs nearest T8. In the longitudinal montage, channels Fp2F8 and F8-T8 show downward deflections because, in each case, G2 of each channel is more negative than G1. The amplitude of the deflection is greater in F8-T8 because the potential is nearer and stronger at that location, whereas channel Fp2-F8, being farther away, records a weaker potential. Conversely, channels T8-P8 and P8-O2, in amplitudes reflecting their distance from the discharge, show upward deflections because G1 of each pair is more negative than G2. This pattern of opposite deflections, channels pointing out-of-phase, is called phase reversal. Phase reversal is the main means of identifying the location of a focal discharge with bipolar montages.

Figure 2-7. Phase reversal is the main means of identifying the location of a focal discharge with bipolar montages.

In the referential montage, the right scalp electrodes are referred to the left ear. All channels deflect upward because the negative potential at T8 renders G1 of each pair more negative than the common reference A1. The amplitude of deflection increases with the proximity of the electrode to the focal discharge. Amplitude is the main means of identifying the location of a focal discharge with referential montages.

24

Acquisition of the Electroencephalogram

Question: Predict the pattern of deflections in the EEG caused by the negative discharge at P8 in Figure 2-8.

Figure 2-8.

Answer: In the bipolar montage, an upward deflection in channel P8-O2 indicates that G1 (P8) is more negative than G2 (O2). Channels T8-P8 and F8-T8 deflect downward because G1 is less negative than G2 (Fig. 2-9). Channel Fp2-F8 is isopotential (no deflection) because both inputs carry the same potential. In the referential montage, the highest upward deflection is at P8, indicating that G1 is strongly more negative than G2. Smaller upward deflections occur in nearby electrodes.

Figure 2-9. Channels T8-P8 and F8-T8 deflect downward because G1 is less negative than G2.

Clinical Pearls 1. with 2. with

Phase-reversal is the main means of identifying the location of a focal discharge bipolar montages. Amplitude is the main means of identifying the location of a focal discharge referential montages.

Localization 1

25

Localization 2

Questions: Continue predicting the pattern of EEG deflections induced by the following discharges. 1. See Figure 2-10.

Figure 2-10.

2.

See Figure 2-11.

Figure 2-11.

26

Acquisition of the Electroencephalogram

3.

See Figure 2-12.

Figure 2-12.

Answers: 1. This example shows a positive potential nearest F8. Note that the direction of deflections for the positive discharge is opposite that of a negative discharge (Fig. 2-13).

Figure 2-13. Note that the direction of deflections for the positive discharge above is opposite that of a negative discharge.

2. This example shows the result of a negative discharge equally spaced between electrodes T8 and P8; channel T8-P8 in the bipolar montage is isopotential. A phase reversal, however, is still present in channels F8-T8 and P8-O2. Therefore, a phase reversal in a bipolar montage need not occur in adjacent channels. The referential montage, reflecting that P8 and O2 are equally involved, shows that the amplitude of the negative discharge is equal in both T8-A2 and P8-A2 (Fig. 2-14).

Localization 2

27

Figure 2-14. The referential montage, reflecting that P8 and O2 are equally involved, shows that the amplitude of the negative discharge is equal in both T8-A2 and P8-A2.

3. Figure 2-12 shows a negative discharge that occurs somewhat posterior to electrode O2, a location termed end-of-chain. In an end-of-chain discharge, the expected phase reversal in the bipolar montage does not occur because there is no electrode on the other side to straddle over the focal potential. Therefore, each G1-G2 pair records that G2 is more negative than G1. Each channel, as a result, deflects downward. On the other hand, in the referential montage G1 from each pair is more negative than G2, and each channel deflects upward in proportion to its distance from the occipital potential (Fig. 2-15).

Figure 2-15. In the referential montage G1 from each pair is more negative than G2, and each channel deflects upward in proportion to its distance from the occipital potential.

Clinical Pearls 1. Isopotential channels can result from either being equally uninvolved by a focal discharge or from being equally involved. 2. Focal discharges that occur at the end-of-chain may show no phase reversal on a bipolar montage.

28

Acquisition of the Electroencephalogram

Localization 3

Questions: Continue predicting the pattern of EEG deflections induced by the following discharges. 1. See Figure 2-16.

Figure 2-16.

2.

See Figure 2-17.

Figure 2-17.

Localization 3

29

3.

See Figure 2-18.

Figure 2-18.

Answers: 1. This example shows a negative discharge that is isopotential between T8 and A2. Note that phase reversals appear in the referential montage. A contaminated reference occurs when the discharge of interest involves the reference electrode. Because channels share a common reference, all channels display the discharge, thereby defeating the localizing abilities and usefulness of the selected referential montage. In this case, selection of an uninvolved reference, such as the contralateral ear, will avoid contamination (Fig. 2-19).

Figure 2-19. Selection of an uninvolved reference, such as the contralateral ear, will avoid contamination.

2. This example shows the result of a horizontal dipole. Only one half of a discharge is usually visible on the upper hemisphere of the scalp recording; the other half of the dipole points down into the inaccessible hemisphere of the scalp model. In some cases, however, dipoles can occur horizontally, thereby exposing both the positive and negative ends of the electrical dipole to regions that are accessible to scalp recordings. In both Figures A and B, phase reversals are apparent in referential montages. The only time in which phase reversals occur in referential montages are in the case of a contaminated reference or a horizontal dipole (Fig. 2-20).

30

Acquisition of the Electroencephalogram

Figure 2-20. Phase reversals occur in referential montages, only in cases of a contaminated reference or a horizontal dipole.

3. This example shows what may occur in a broadly transmitted or a generalized discharge, with a series of unpredictable or isopotential discharges present in bipolar montages. In these cases, the region of highest amplitude on the referential montage identifies the probable source of the discharge (Fig. 2-21).

Figure 2-21. discharge.

The region of highest amplitude on the referential montage identifies the probable source of the

Clinical Pearls 1. Phase reversals occur in referential montages because of either contaminated references or horizontal dipoles. 2. Focal discharges with a broad field of distribution and generalized discharges are best interpreted with the use of a referential montage.

Localization 3

31

Calibration and Technical Requirements

The minimal technical requirements of clinical EEG are designed to ensure uniformity in acquisition and to provide a study that is reproducibly interpretable. The American Clinical Neurophysiology Society (ne´e American EEG Society) specifies minimum criteria for routine and neonatal EEG as well as other neurophysiologic tests. An EEG should consist of a minimum of 16 cerebral channels displayed in at least three montages during the course of the study—a longitudinal bipolar, a transverse bipolar, and a referential montage. The minimum study duration for routine studies is 20 minutes and 1 hour for neonates. Many requirements are plain common sense. EEG studies must contain name, age, and other identification so that the tracing and its report can be easily linked. Channels and changes to sensitivities and filters must be labeled after every change. Patient movement and state must be commented upon. Required calibration demonstrates that all channels display EEG voltages as accurately as possible. Calibration consists of three parts: prestudy machine calibration, biocalibration, and poststudy machine calibration. In traditional paper-based systems, machine calibration consists of the application of a calibration signal, usually a 50-mV square wave, into all inputs. The alternating ‘‘shark-fin’’ appearance of the output waveform results from the application of a filter (Fig. 2-22). Machine calibration allows checking whether paper transport is perpendicular to the line of pens, whether pens are aligned, whether pens are equally damped (writing upon the paper with equal force and ease of movement), whether ink flow is smooth, whether filters are applied equally, and whether amplifiers cause equal pen swings for the identical voltage input.

Figure 2-22. During machine calibration (top), the alternating ‘‘shark-fin’’ appearance of the output waveform results from the application of a filter. During biocalibration (bottom), all channels display the same input from the scalp.

32

Acquisition of the Electroencephalogram

Biocalibration uses the subject as the signal source, not only double-checking the quality factors mentioned earlier, but also ensuring that the tracing indeed reflects biologically relevant signal. In biocalibration, all channels consist of identical inputs Fp1 to the G1 input and O2 to G2. The montage was designed to amplify biologic signal as much as possible (by virtue of the long interelectrode distance) and to demonstrate the most quickly recognized EEG findings in awake individuals, the posterior waking rhythm, and eye movement artifact. Finally, a second set of calibrations using the 50-mV calibration signal is performed at the study’s end to demonstrate the combinations of filter settings and sensitivities used during the study. Digital EEG systems vary by manufacturer in the exact method of machine calibration. Some manufacturers dispense with display of calibration and perform calibration internally. For example, at the start of the study, a sinusoidal calibration signal is sent to all amplifiers. The EEG acquisition program reads the output and creates a calibration table that contains a correction factor by which each actual output is multiplied. Each channel is adjusted mathematically to display equal output. Questions: 1. What is the minimum duration or recording? 2. What montages are required? 3. What channels does standard biocalibration use? Answers: 1. 20 minutes. 2. Required montages are longitudinal bipolar, transverse bipolar, and one referential. 3. Fp1-O2.

Clinical Pearls 1. Minimum technical criteria for performance of EEGs exist to maintain quality and consistency from lab-to-lab and study-to-study. 2. Although techniques differ by recording method (analog or digital), all EEG studies require machine and biocalibration. 3. Minimum requirements for routine EEG include 16 channels, 3 montages, and 20 minutes duration. REFERENCE 1. American Electroencephalographic Society: Guideline one: Minimum Technical Requirements for Performing Clinical Electroencephalography. J Clin Neurophysiol 1994; 11:2–5.

Calibration and Technical Requirements

33

The Electro-oculograph (EOG)

Eye leads, once used mainly in polysomnographic studies, are now practical to use in routine EEG recordings. Eye leads can help differentiate anteriorly dominant potentials that originate from eye movement versus cerebral potentials and other noncerebral artifact. The neural tissue of the retina that lines the globe maintains a negative potential relative to the scalp. The corneal and pupil, however, are positive by virtue of the absence of neurons. Therefore, the anterior portion of the globe is relatively positive to the posterior portion. ‘‘Nerve Negative, Pupil Positive’’ is a mnemonic useful in determining the pattern of potentials recorded with eye leads. Figure 2-23 shows how eye leads channels are created in the examples in this book. Although there are various ways to create eye lead channels, the common feature among them is placement of the paired eye electrodes at opposing elevations—one supraorbital and one infraorbital, with both placed laterally to the eye. This book designates such electrodes as ‘‘E1’’ and ‘‘E2’’; some labs use ‘‘LOC’’ and ‘‘ROC’’ for left and right outer canthus.

Figure 2-23. Placement of eye lead channels.

The construction of channels E1-A1 and E2-A2 guarantees that every movement of the eyes will cause a corresponding phase reversal across the two channels. An upward glance causes the left pupil to move toward E1 and record a relative positivity in channel E1-A1. The right pupil simultaneously moves away from E2, and channel E2-A2 records a relative negative potential. Similarly, a downward glance moves the left pupil away from E1 and the right pupil toward E2. Eye movements, often resembling slow wave activity, sometimes must be differentiated from cerebral anterior activity, without benefit of eye leads in patients who cannot tolerate their application. Even with eye leads, some patients have such continuous and intrusive eye movement that it obscures frontal activities. Technologists in this case may help by recording a portion of the study while holding eyes closed.

34

Acquisition of the Electroencephalogram

Question: The patient in this study was asked to look to either side. To which side did the patient look at point A and to which side at B in Figure 2-24?

Figure 2-24.

Answer: At A the patient looks rightward; at B, leftward. As Fig. 2-25 shows, at A E1-A1 records upward deflections indicating a negativity, and E2-A2 records downward deflections indicating a positivity. The lateral eye movement that can cause this pattern is a rightward glance as the negative nerve swings toward E2 and the positive pupil toward E1. The opposite occurs at B. Note that the high sensitivities used to demonstrate pulse reversal also cause pen blocking.

Figure 2-25. A, E1-A1 records upward deflections, indicating a negativity, and E2-A2 records downward deflections, indicating a positivity. The lateral eye movement that can cause this pattern is a rightward glance as the negative nerve swings toward E2 and the positive pupil toward E1. B, The opposite occurs.

Clinical Pearls 1. The globe is relatively positive anteriorly and negative posteriorly. 2. Eye movement artifact causes phase reversals in properly applied eye electrodes. 3. Eye lead channels can distinguish between potentials that appear on the scalp that arise from eye movement artifact versus potentials of other origins.

The Electro-oculograph (EOG)

35

VOCABULARY OF ELECTROENCEPHALOGRAPHIC FINDINGS 3

Electroencephalographic Description

1

Features of an electroencephalographic (EEG) tracing are described in the following terms: Frequency. Actual frequency or frequency bands. Some prefer to use units of cycles per second (cps) when referring to biologic activity and units of Hz when referring to nonbiologic activity. delta 8 cps by age 3. Many studies have determined the normal frequency for age. The plot shown below is a composite of three references. The two older studies show upper and lower limits of a cohort of normal children, most studied longitudinally. The more recent study shows the upper and lower 95% confidence limits of frequencies of normal children. Note that the increase in frequency is rapid through the first year of life and gradually slows thereafter. No pathologic conditions are associated with alpha rhythm frequencies that are faster than the normal range, although care should be taken that the observed supranormal frequencies are the posterior waking rhythm. Frequencies slower than the 95% confidence interval, if drowsy state is excluded, should be considered an indicator of a nonspecific encephalopathy.

The Normal Waking Electroencephalogram

Determining the frequency of alpha rhythm in infants and small children can be difficult because, when fully awake, they often maintain eye opening and block alpha rhythm, leaving behind other activities that can mistakenly be counted as alpha rhythm. Furthermore, when alpha rhythm can usually be best seen in these young subjects—quietly restful with eyes closed to elicit a and approaching sleep—the alpha rhythm can be slowed from drowsiness. The solution is to have the EEG technologist hold the eyes closed to elicit a sample of waking alpha rhythm. Also adding to the difficulty is that, in children, occipital slow wave discharges are often normally superimposed upon faster occipital frequencies, as seen in this case. The lower limit of Lindsey’s study of the longitudinal development of the posterior rhythm probably reflects this intermixing of posterior slow waves with the best ‘‘alpha rhythm’’ frequency. Current technique is to count the best frequency of the posterior rhythm, as reflected in Smith’s study of alpha rhythm development.

51

Clinical Pearls 1. Alpha rhythm is first seen at age 3 months and increases with maturation. 2. Alpha rhythm frequency of >8 cps is attained around age 3. 3. Explicit demonstration of alpha rhythm in children is required in most studies of those too young to cooperate with instructions. REFERENCES 1. Eeg-Olofsson O, Petersen I: The development of the EEG in normal children from the age of 1 to 15 years: Paroxysmal activity. Neuropadiatrie 1971; 4:375–404. 2. Lindsley DB: Longitudinal study of the occipital alpha rhythm in normal children: Frequency amplitude standards. J Genet Psychol 1939; 55:197–213. 3. Smith JR: The electroencephalograph during normal infancy childhood. I: Rhythmic activities present in the neonate and their subsequent development. J Genet Psychol 1938; 53:431–453.

52

The Normal Waking Electroencephalogram

PATIENT 4 A 6-year-old girl with seizures after a urinary tract infection

A 6-year-old girl, seizure-free for 3 years, is evaluated after recurrence of generalized tonicclonic seizures during a severe urinary tract infection. She takes topiramate and felbamate. The tracing below was recorded with the patient awake. Question:

Is the posterior rhythm normal for age? Note the transverse longitudinal montage.

The Normal Waking Electroencephalogram

53

Answer: Sporadic delta activity (arrows) accompanies posterior alpha activity in the 10–10.5 cps range and appears with eye closure. Posterior slow waves of youth are an age-dependent normal finding. The alpha rhythm is normal for age. Discussion: One difficulty the waking EEG of a child presents is the accurate distinction between normal and abnormal occipital delta activity. Posterior slow waves of youth are most likely to occur in children between ages 8 and 14 and are rare beyond age 21. Posterior slow waves of youth present as polyphasic, sporadic, or arrhythmic occipital delta activities that otherwise behave like the alpha rhythm. They appear maximally during relaxed wakefulness and disappear with drowsiness or sleep. Like alpha

rhythm, posterior slow waves of youth are frequently higher in amplitude on the right side. Alpha rhythm in the alpha frequency band is typically superimposed upon the higher amplitude slow wave discharges, giving the occipital alpha rhythm a poorly regulated appearance. Most importantly, similar to alpha rhythm, posterior slow waves of youth attenuate with eye opening. In distinction to posterior slow waves of youth, unreactive or rhythmic delta activity that interrupts alpha rhythm is abnormal.

Clinical Pearls 1. Posterior low waves of youth are a frequent finding in normal children. 2. Posterior slow waves of youth are distinguished from abnormal slowing in that they are present with age-appropriate alpha rhythm and, like alpha rhythm, attenuate with eye opening. 3. Reports of abnormal posterior slowing in children should be greeted with skepticism if reactivity, state, and other posterior activities are not adequately described. REFERENCES 1. Aird RB, Gastaut Y: Occipital posterior electroencephalographic rhythms. Electroencephalogr Clin Neurophysiol 1959; 11:637–656. 2. Eeg-Olofsson O, Petersen I: The development of the EEG in normal children from the age of 1 to 15 years: Paroxysmal activity. Neuropadiatrie 1971; 4:375–404. 3. Petersen I, Eeg-Olofsson O: The development of the EEG in normal children from the age of 1 to 15 years: Nonparoxysmal activity. Neuropadiatrie 1971; 2:247–304.

54

The Normal Waking Electroencephalogram

PATIENT 5 A 40-year-old man with depression

A 40-year-old man presents with medically intractable depression before electroconvulsive therapy (ECT). Medications are Zoloft and valproic acid. Question:

What frequency is the alpha rhythm? Is it best counted at sample A or sample B?

The Normal Waking Electroencephalogram

55

Answer: 11.5 cps, sample B. Sample A shows a transiently increased frequency of approximately 12.5 cps after eye closure, called squeak phenomenon. Discussion: The frequency of the alpha rhythm may not be stable throughout a single study. Its frequency can be increased immediately after eye closure. For example, in the tracing the alpha frequency immediately after eye closure artifact at sample A is approximately 12 cps, whereas later in the recording at sample B it is more accurately counted at 11.5 cps. This transient increase in the frequency of alpha rhythm immediately after eye closure is called squeak phenomenon, after the brief harsh squeaking sound made by the old EEG systems that arose from the sound of pens ‘‘squeaking’’ with the abrupt onset of activity after eye closure.

Drowsiness is a state that commonly decreases the frequency of the alpha rhythm. For accuracy’s sake, alpha rhythm should be counted during maximum alertness but not limited to intervals immediately after eye closure. Given these limitations, alpha rhythm should be assigned the highest frequency not increased by squeak phenomenon. In this patient, the alpha rhythm outside of squeak phenomenon was normal. This suggests that his depression is a primary mood disorder rather than one secondary to encephalopathy. Therefore, ECT remains an appropriate treatment.

Clinical Pearls 1. Squeak phenomenon is a transient increase in alpha frequency following eye closure that may cause overestimation of alpha rhythm. 2. Drowsiness can decrease the frequency of alpha rhythm. 3. The ‘‘true’’ alpha rhythm frequency is the fastest frequency during maximum arousal from samples not limited to those immediately after eye closure.

56

The Normal Waking Electroencephalogram

PATIENT 6 A 25-year-old man with spells of loss of consciousness

A 25-year-old man is evaluated for recent onset of spells of loss of consciousness and wandering. He is on no medications. The tracing below was recorded with the patient awake. Question:

Is the posterior rhythm before eye opening (EO) normal?

The Normal Waking Electroencephalogram

57

Answer: Yes. Approximately 20 cps posterior rhythm admixed with 10 cps posterior rhythm is an example of fast alpha rhythm variant, a normal finding. Discussion: In some recordings, the 8–13 cps frequency of the alpha rhythm is infrequent, even in relaxed, normal subjects. Instead, the majority of the posterior waking rhythm consists of low amplitude (25 cps likely result from EMG activity and can obscure interpretation of the tracing. 2. Technologists may improve artifact by patient manipulation. Sleep or sedation may be required in uncooperative subjects. 3. EEG reports must state that the recording is normal, abnormal, essentially normal, or technically insufficient.

The Normal Waking Electroencephalogram

69

PATIENT 12 A 51-year-old woman with spells of diaphoresis and unresponsiveness

A 51-year-old woman is evaluated for risk of epileptic seizures after she had several spells of diaphoresis, tachypnea, and subsequent unresponsiveness. She is taking no medications. The patient was instructed to stay up all night before the recording. The EEG was recorded with the patient awake. Question: What are the posterior sharp transients called that occur after eye opening (arrows at enlarged inset)?

70

The Normal Waking Electroencephalogram

Answer: Sporadic positive waves present in occipital channels during eye opening are called lambda waves and are findings designating saccades. Discussion: In contrast to eye movements that generate artifact, lambda waves are cerebral activities that correspond to the saccades that accompany visual fixation and scanning. They usually appear as irregular trains of biphasic, sharp, surface-positive potentials in occipital channels. Although, in the example, they do not exceed 30 mV, they can be quite prominent in children. Lambda waves can also be polyphasic with a prominent negative deflection. The appearance of lambda waves decreases with increasing age. Brightly illuminated rooms can exacerbate their appearance. The technologist can differentiate them from pathologic sharp waves by demonstrating their dependence on bright room lighting or their appearance with visual scanning. In this particular case, the recording was performed following partial sleep deprivation. Sleep deprivation serves three purposes. First, sleep deprivation is an accepted activation procedure. Activation procedures attempt to increase the yield of IEDs in susceptible individuals. Most studies have shown that yield of IEDs increases following sleep deprivation. Second, sleep deprivation promotes the occurrence of sleep. Sleep itself is an important activation procedure, because certain epilepsy syndromes have IEDs that appear more frequently during certain sleep-wake states. Sleep and

The Normal Waking Electroencephalogram

sleep deprivation have cumulative effects in increasing the yield of IEDs. Sleep also improves the quality of recordings in lieu of sedation. Chloral hydrate was the standard agent for sedation of uncooperative patients during EEG, largely for the belief that it had little effect on normal activities or on the incidence of IEDs. Studies show no clear morbidity related to use of chloral hydrate for pediatric sedation in EEG, but cases of mortality or morbidity have been reported for overdoses. Some laboratories encourage use of mild sedatives, such as melatonin or diphenhydramine, instead of chloral hydrate. Such ‘‘conscious sedation’’ requires the training of staff in resuscitation and monitoring of vital signs, or scheduling physician-supervised anesthesia during procedures. Heightened concerns over medical liability and increasing requirements for specialized training and monitoring make routine sedation impractical for most laboratories. This shift in practice appears for the better, with some authorities reporting no clear changes in the rates in unsuccessful tests, despite a drastic drop in sedation use. Sleep deprivation, sleep, and judicious scheduling of potentially uncooperative patients in the afternoon (when circadian cycles and sleep deprivation make it a time of day especially favorable to sleep) have supplanted pharmacologic sedation in many U.S. laboratories.

71

Clinical Pearls 1. Lambda waves are sharp, transient, surface positive discharges accompanying saccadic eye movement during visual scanning. 2. As we will see in subsequent cases, ‘‘sharpness’’ is not synonymous with epilepsy. Benign findings can have a sharp morphology. 3. Sleep and sleep deprivation can activate the appearance of IEDs. 4. Sleep and sleep deprivation are important replacements for sedation once routinely used for recording EEGs in children and uncooperative patients. REFERENCES 1. Cote CJ, Karl HW, Notterman DA, et al: Adverse sedation events in pediatrics: Analysis of medications used for sedation. Pediatrics 2000; 106:633–644. 2. Fountain NB, Kim JS, Lee SI: Sleep deprivation activates epileptiform discharges independent of the activating effects of sleep. J Clin Neurophysiol 1998; 15(1):69–75. 3. Olson DM, Sheehan MG, Thompson W, et al: Sedation of children for electroencephalograms. Pediatrics 2001; 108:163– 165. 4. Roth M, Green J: The lambda wave as a normal physiological phenomenon in the human electroencephalogram. Nature 1953; 172:864–866. 5. Thoresen M, Henriksen O, Wannag E, et al: Does a sedative dose of chloral hydrate modify the EEG of children with epilepsy?. Electroencephalogr Clin Neurophysiol 1997; 102:152–157.

72

The Normal Waking Electroencephalogram

PATIENT 13 A 4-year-old girl with headaches and inattention

A 4-year-old girl has been sent home on numerous occasions for diffuse headaches and inattention. The referring physician is considering absence seizures. The child is awake, hyperventilating, and taking no medications. The child initially refuses hyperventilation but successfully performs when asked to blow on a pinwheel. Eye leads are omitted because of poor cooperation. Question:

What is the finding in the following tracing?

The Normal Waking Electroencephalogram

73

Answer: The recording shows a burst of diffusely distributed, rhythmic, high-amplitude 2 cps delta activity induced by hyperventilation. This response is normal ‘‘buildup.’’ Discussion: Hyperventilation for one or two 3- to 4-minute trials is a routine activation procedure. The term buildup is the term for the slowing that emerges from central nervous system (CNS) blood flow changes induced by hypocapnia. In children, buildup often takes the form of high-amplitude, generalized, synchronous, symmetric rhythmic delta activity. Less impressive responses, such as emergence of low-amplitude bursts of symmetric theta activity, can be present in older children or adults. Adults often have no discernible changes. Seizures in susceptible patients are the only unequivocal abnormalities induced by hyperventilation. Other findings during hyperventilation must be interpreted in context.

Asymmetric or focal slowing limited to hyperventilation is a relative abnormality that indicates subtle dysfunction over the side with more severe slowing. Persistent buildup, that which lasts over a minute following cessation of hyperventilation, is thought by some to be a nonspecific abnormality, but persistent slowing or marked buildup can result from hypoglycemia. The EEG technologist should note routinely the time of the patient’s last meal, and, if buildup is remarkable in duration, a retrial is often performed after administration of a glucose-containing snack. Hyperventilation is avoided in those with cardiovascular or neurovascular disease because vasoconstriction is a possible result of hypocapnia.

Clinical Pearls 1. Hyperventilation is a routine activation procedure that is most useful in provoking absence seizures. 2. Buildup is the symmetric slowing induced by hyperventilation-induced hypocapnia.

74

The Normal Waking Electroencephalogram

PATIENT 14 A 14-year-old girl with spells of headaches and confusion

A 14-year-old girl has headaches and confusion thought to be consistent with atypical migraine or complex partial seizures. The recording was performed with the patient awake and on no medications. This portion is recorded during intermittent photic stimulation, with each flash of a strobe light indicated by the tic marks in the photic channel. Question: indicate?

What does the response in the tracing shown below, during photic stimulation,

The Normal Waking Electroencephalogram

75

Answer: The burst of occipital activity following the frequency of the photic stimulation is a normal driving response to photic driving. Discussion: Intermittent photic stimulation is a standard activation procedure. Although various protocols are used, usually patients are exposed to a series of 10-second blocks of stimulation at frequencies starting at 2 Hz and increasing to 20 Hz, before decreasing again to 2 Hz. Many centers repeat stimulation with eyes open and closed. Others expose patients to blocks of gradually and continuously increasing or decreasing flash frequencies. The normal response to photic stimulation is a symmetric driving response, a rhythmic occipital or posteriorly dominant activity that occurs at the primary frequency or at a slower harmonic to the rate of the flashes. Sometimes a brief burst of sharp activity with onset or offset of stimulation occurs, a normal ‘‘onresponse’’ or ‘‘off-response.’’ Subharmonic driving and an on-response are shown on the second example taken from another patient. Driving responses need not be continuous throughout the stimulation. Indeed, the ‘‘onresponse’’ in this example (as well as other published examples) may consist of a fragment of rhythmic photic driving.

76

Another normal response is the lack of photic driving. Photomyoclonus (photomyogenic) responses are the twitching of facial and sometimes upper trunk muscles in response to each flash. No epileptic discharges accompany photomyoclonus, but anterior muscle artifacts may be present. It is best thought of as an exaggerated startle response. Some observe that photomyoclonus may appear during the acute phase of alcohol withdrawal and may be evidence of sympathetic hyperactivity. The only unequivocally abnormal response to intermittent photic stimulation is a triggered epileptic seizure. A more common abnormal finding is a photoparoxysmal response, a burst of asymptomatic generalized multiple spike-wave discharges. Other relative abnormalities include asymmetry, during which driving responses are present on one side and absent on the other. Its significance, however, is limited in the absence of corroborating focal abnormalities. High-amplitude responses or high-amplitude occipital spikes are seen in neuronal ceroid lipofuscinosis.

The Normal Waking Electroencephalogram

Clinical Pearls 1. Intermittent photic stimulation is an activation procedure intended to induce interictal epileptiform discharges or seizures in susceptible individuals. 2. Normal results of intermittent photic stimulation include symmetric photic driving or lack of responses. 3. Photomyoclonus is an exaggerated, nonepileptic motor response to photic stimulation. 4. Photoparoxysmal responses consist of generalized polyspike-wave discharges in response to photic driving. REFERENCE 1. Trenite DG, Binnie CD, Harding GF, et al: Medical technology assessment photic stimulation—Standardization of screening methods. Neurophysiol Clin 1999; 29:318–324.

The Normal Waking Electroencephalogram

77

THE NORMAL ELECTROENCEPHALOGRAM OF SLEEP 5

PATIENT 15 A 23-year-old woman undergoing overnight video-EEG recordings (Part 1)

A 23-year-old woman undergoes overnight video-EEG recordings for diagnosis of spells of unknown etiology. A baseline recording while awake is normal, featuring a 10-cps alpha rhythm. The patient is taking no medications. Questions: What is the source of the phase-reversing slow waves seen best at F7 and F8 (arrows)? In what state is this patient?

79

Answers: The source of the slow waves is lateral eye movements. The patient is in a state of drowsiness (stage 1 sleep). Discussion: Polysomnography (PSG) is required for formal sleep scoring. PSG, with the use of central and occipital EEG electrodes, eye movement leads, and EMG leads, assigns one of six sleep stages to 30-second epochs of data. These stages are as follows:

Several features identify the adult drowsy state in the previous tracing. Slowing or attenuation of the alpha rhythm from the patient’s normal waking posterior rhythm denotes drowsiness. In some patients, the frequency gradually slows; in others, it abruptly attenuates and is replaced by theta activity. Emergence of diffusely distributed theta activCommon Predominant ities. In formal sleep scoring, more than 50% of a 30-second epoch must consist of theta activity. EEG Eye Slow lateral eye movements, seen as ~0.5 cps Stage Name Frequency Movement EMG slow waves in the anterior channels of this examStage W Wakeful- Alpha þþ þþ ple, commonly occur with the transition to and the onset of sleep. ness 1 Drowsi- Theta slow rolling þ Note that no eye leads were used on this study. ness Because most patients remain on intensive moni2 Light Theta  þ toring for several days instead of the standard sleep 8 hours of routine PSG, eye leads are used only 3 and 4 Deep Delta  þ if there are particular questions to answer. Even sleep then, facial skin is sensitive, and eye leads are only REM REM Alpha/theta þþ – placed short term. Lateral eye movements usually sleep cause phase reversals best seen in electrodes F7 or REM, rapid eye movement. F8, helpful evidence when eye leads are absent. Other findings common in drowsiness include Stages 1–4 are often grouped together as non- transient enhancement of beta activity. HypersynREM sleep. During a normal duration of noctur- chrony (brief bursts of generalized delta activity nal sleep, most healthy adults spend about 75% of during light sleep) are often present in children time in non-REM sleep and 25% in REM sleep. and young adults. Benign interictal epileptiform Sleep stages typically occur in brief episodes tied discharges—sharp waves without pathologic to a 90- to 110-minute ultradian (less than 1 day) correlates—often appear during drowsiness or light sleep. Bursts of theta activities, usually cycle. Although formal sleep scoring is inappropriate to generalized but sometimes with a left or right undertake on the data usually acquired with clinical hemispheric prominence, occur from time to time. Drowsiness is distinguished from mild encephEEG, it is important to recognize characteristic changes of state so that they can be distinguished alopathy by its transient nature; the EEG technolfrom pathologic changes. In fact, to interpret EEG, ogist should stimulate the patient to demonstrate the interpreter must know unambiguously what arousal (and normal waking activities) to provide clinical state the patient is in during the recording. a comparison.

Clinical Pearls 1. Drowsiness (stage 1 sleep) is marked by attenuation or slowing of alpha rhythm, emergence of diffusely distributed theta activities, and, in some cases, slow rolling, lateral eye movements and enhancement of beta activities. 2. Arousal and return to normal waking activities distinguishes normal drowsiness from mild encephalopathy. REFERENCE 1. Rechtschaffen A, Kales A: A Manual of Standardized Terminology, Techniques, and Scoring System for Sleep Stages of Human Subjects. Los Angeles, Brain Information Service, Brain Research Institute, UCLA, 1968.

80

The Normal Electroencephalogram of Sleep

PATIENT 16 A 23-year-old woman undergoing overnight video-EEG recordings (Part 2)

This excerpt from the same patient is taken 5 minutes after the previous sample. Questions: Name the diffusely distributed findings at times a and b and the occipital transients at c. In what state is the patient?

The Normal Electroencephalogram of Sleep

81

Answers: (a) K complex; (b) sleep spindle; (c) positive occipital sharp transients of sleep. The patient is in light sleep (stage 2 sleep). Discussion: The predominant frequency during stage 2 sleep (light sleep) is theta activity, although delta activities may appear intermittently. Stage 2 sleep is defined as the first appearance of sleep spindles. Sleep spindles are symmetric, synchronous bursts of 11–14 cps activities, with maximal amplitudes in central regions. Durations in adults typically remain between 1 and 3 seconds. Sleep spindles have characteristic spindle-form morphology with amplitudes at the beginning and end of each burst being smaller than the midportion. Although their appearance defines stage 2 sleep, they often appear in deeper stages of non-REM sleep but are harder to see because of state-related changes in background activity. K complexes are high-amplitude, biphasic or polyphasic centrally dominant discharges that

emerge during stage 2 sleep. Sleep spindles can appear at the end of a K complex, as seen in this case. Vertex sharp transients, another characteristic midline finding during sleep, can also appear near or linked to K complexes. K complexes are frequently evoked by auditory stimulation, such as noise in the hallway outside of the EEG recording room. Positive occipital sharp transients of sleep (POSTS) may occur during stage 1 or 2 sleep. They appear most commonly in young adults but are not limited to this age group. Although they appear synchronously in occipital channels, the amplitudes are often asymmetric. Their importance lies in their morphology; they can be mistaken for pathologic sharp waves.

Clinical Pearls 1. Sleep spindles are spindle-form, synchronous, centrally dominant bursts of high alpha- or low-frequency beta activity. Their appearance defines stage 2 sleep. 2. K complexes are high amplitude, biphasic or polyphasic midline discharges that appear during light sleep and can be evoked by auditory stimuli. 3. POSTS are bisynchronous, usually asymmetric, sharp transients that have a positive potential recorded from occipital regions. REFERENCE 1. Loomis AL, Harvey EN, Hobart G: Distribution of disturbance patterns in the human electroencephalogram with special reference to sleep. J Neurophysiol 1938; 1:413–430.

82

The Normal Electroencephalogram of Sleep

PATIENT 17 A 23-year-old woman undergoing overnight video-EEG recordings (Part 3)

This excerpt, taken from the same monitoring session as the previous two samples, is recorded 15 minutes after the first excerpt. Question:

In what state is this patient?

The Normal Electroencephalogram of Sleep

83

Answer:

Deep sleep (stages 3–4 sleep).

Discussion: The predominance of delta activities defines sleep stages 3 and 4 (deep sleep, slow wave sleep). Delta activities in stages 3–4 sleep appear as semirhythmic, generalized delta activities with amplitudes ¼ 75 mV. Sleep spindles, K complexes, vertex sharp transients, and other mixed frequencies can also be present but may be less easily distinguishable from highamplitude delta activities. Formal sleep scoring with the use of PSG restricts delta activity to those activities ¼ 2 cps that ¼ 75 mV in amplitude obtained in specific central and occipital channels referred to contra-

lateral references (C3-A2, O1-A2, C4-A1, O2A1). Stage 3 is defined when delta activity occupies 20–50% of a 30-second epoch and stage 4 more than 50% of an epoch. Because most clinical EEG is performed during the day, and because sleep usually starts with stage 1 sleep, deep sleep is rarely seen in the daytime realm of the typical EEG lab. Slow wave sleep is commonplace, however, during overnight recordings in epilepsy monitoring units. The recognition of deep sleep is important because delta activities may be present in cases of moderate to severe encephalopathy.

Clinical Pearls 1. Stages 3 and 4 are defined as the appearance of ¼ 75 mV, semirhythmic generalized delta activities during sleep. 2. Sleep scoring uses a more restrictive definition of the delta frequency band (2 cps ¼ 75 mV) than clinical EEG (50%.

84

The Normal Electroencephalogram of Sleep

PATIENT 18 A 23-year-old woman undergoing overnight video-EEG recordings (Part 4)

This excerpt from the same patient is taken another 20 minutes after the previous sample. The patient is taking no medications. Questions:

Name the findings marked at the arrows. In what state is this patient?

The Normal Electroencephalogram of Sleep

85

Answers:

The finding marked at the arrows is REMs. The patient is in REM sleep.

Discussion: REM sleep is formally defined as a state during which rapid eye movements appear during EEG activities of low amplitude, asynchronous, generalized mixed alpha and theta activities, and during hypotonia measured by submental EMG. REMs are quick, saccade-like movements, unlike the slow lateral movements seen in drowsiness. They typically appear in bursts and, conversely, may be absent for long stretches of REM sleep, requiring the formulation of so-called ‘‘REM rules’’ in sleep scoring to increase consistency. REM hypotonia, determined by the lack of muscle activity in submental EMG leads, is

present and important in distinction of REM from light sleep or wakefulness. REM sleep is sometimes called paradoxical sleep because the desynchronized, low-amplitude, mixed alpha and theta activities appear more similar to those of wakefulness than the slower, more synchronized activities of non-REM sleep. Like slow-wave sleep, REM sleep is seldom encountered in standard clinical EEG laboratory. The mean latency of onset of REM exceeds 20 minutes in healthy adults. Sleep onset REM, however, can be seen after severe sleep deprivation or in primary sleep disorders, such as narcolepsy.

Clinical Pearl REM sleep is defined by the appearance of REMs, hypotonia, and desynchronized, low-amplitude, mixed theta, and alpha frequency activities. REFERENCE 1. Rechtschaffen A, Kales A: A Manual of Standardized Terminology, Techniques, and Scoring System for Sleep Stages of Human Subjects. Los Angeles, Brain Information Service, Brain Research Institute, UCLA, 1968.

86

The Normal Electroencephalogram of Sleep

PATIENT 19 A 30-year-old woman with depression with psychotic features

An EEG is requested to evaluate possible encephalopathy in a 30-year-old woman with major depression with psychotic features, who was scheduled for electroconvulsive therapy (ECT). Medications were haloperidol and mirtazapine. The recording is made with the patient asleep. Questions: Do diffuse beta activities on the early portion of the sample (under bar) indicate medication effects? What is the transient present at Cz at the arrow?

The Normal Electroencephalogram of Sleep

87

Answers: sleep.

Beta activities of drowsiness are normal. Present at Cz is a vertex sharp transient of

Discussion: Beta activities may become transiently enhanced during drowsiness or light sleep. They should be symmetric and diffusely distributed. Enhanced beta activities, because of drug ingestion, on the other hand, should be present throughout the tracing regardless of state. The EEG is sensitive in detection of encephalopathies—states of altered consciousness or delirium. Psychiatric diseases largely spare the level of consciousness. Thus, the main use of EEG in the evaluation of psychiatric diseases is to evaluate possible encephalopathy that may be difficult to distinguish from primary psychiatric disease. In the case of treatment of depression with ECT, EEG may aid in screening of organic causes of depression that may not respond to ECT. Following ECT, EEG may aid in determining whether clinical responses stem from changes in mood versus postictal confusion or complications, such as ECT-induced nonconvulsive status epilepticus. Some psychiatric disorders may be associated with epileptiform abnormalities. Various studies

find that benign epileptiform variants are often seen in patients with schizophrenia or depression. Other studies have attempted to use digital EEG techniques to map out the distributions of the various frequency bands that may be characteristic of certain psychiatric diseases. Critics of these studies point out that statistical techniques, data and subject selection, and other experimental problems make these studies controversial. It is clear, however, that the broad overlap of EEG findings between psychiatric patients and controls and within groups of psychiatric patients renders specificity and sensitivity of routine EEG or digital EEG techniques too low for diagnostic use in the evaluation of specific psychiatric diseases. The current patient has a normal EEG, indicating to her psychiatrists that her apparent lethargy, poor responsiveness, and hallucinations are more likely from a psychiatric process rather than one originating from an unappreciated organic brain syndrome.

Clinical Pearls 1. Beta activities may be briefly enhanced in amplitude during drowsiness or light sleep. 2. Beta activities must be interpreted in context of the patient’s state. 3. The main use of EEG in evaluation of psychiatric disease is its ability to evaluate possible organic brain syndromes that present with psychiatric symptoms. 4. No commonly accepted, specific EEG findings are attributed to psychiatric disease. REFERENCES 1. Hughes J: A review of the usefulness of the standard EEG in psychiatry. Clin Electroencephalogr 1996; 27(1): 35–39. 2. Inui K, Motomura E, Okushima R, et al: Electroencephalographic findings in patients with DSM-IV mood disorder, schizophrenia, other psychotic disorders. Biol Psychiatry 1998; 43(1): 69–75. 3. John ER, Prichep LS, Fridman J, Easton P: Neurometrics: Computer-assisted differential diagnosis of brain dysfunctions. Science 1998; 239(4836): 162–169. 4. Oken B, Chiappa K: Statistical issues concerning computerized analysis of brainwave topography. Ann Neurol 1986; 19: 493–494.

88

The Normal Electroencephalogram of Sleep

PATIENT 20 A 22-year-old man with intermittent confusion

A 22-year-old man presents with intermittent confusion. The patient takes olanzapine. The EEG is performed after the patient has improved clinically. He is asleep for most of the recording. Question: and c?

What are the sources and clinical significance of sharp transients at points a, b,

The Normal Electroencephalogram of Sleep

89

Answer: (a) electrode pop at F7, (b) vertex sharp transients of sleep, (c) POSTS. None of these findings is abnormal. Discussion: Distinctions among IEDs, sharp artifacts, and normal sharp transients are some of the main challenges of EEG interpretation. Electrode pop. Electrode pops are transient capacitive discharges caused by abnormalities of the electrode-scalp interface. Pops are easily identified because of their short duration (50% of a 1-minute epoch consists of higheramplitude activities, some designate the background as semidiscontinuous. The duration between bursts decreases with ECA. Discontinuous activities are seen in premature infants (trace´ discontinu [TD]) and persist as a part of quiet sleep in more mature infants (trace´ alternant [TA]). Continuous activities are mixed-frequency activities that do not feature the abrupt changes of amplitude of discontinuous activities. Continuous activities emerge reliably at ECA 30 weeks and become the main background activities of wakefulness and active sleep. Continuous activities of wakefulness and active sleep are often classified further, but the nomenclature is inconsistent. Activite´ moyenne, mixed activities, low voltage irregular (LVI), and trace´ continu all refer to low-amplitude, rather featureless, mixed activities with varying predominance of delta, theta, and alpha frequency activities. In this text, those described previously will be referred to simply as continuous activities. A final subtype of continuous activities consists of semirhythmic, medium- to high-amplitude delta activities termed high voltage slow (HVS) or trace´ continu lentement), which are seen in quiet sleep of full-term neonates. Various physiologic EEG patterns appear at different steps of maturation, help in confirming ECA, and aid in differentiating background activities. In Figure 6-1, the first appearance, maximal expression, and resolution of specific patterns are represented by the gray polygons within each vertical column. Discordance between ECA and expected EEG findings that exceeds 2 weeks is a finding termed dysmaturity. Dysmaturity indicates encephalopathy or serves as a warning that the ECA was misstated. Figure 6-1 details various developmental stages, which are reviewed in subsequent cases. Question: Why are background EEG activities often designated in French? Answer: French investigators, headed by C. Dreyfus-Brisac and colleagues (see following reference for an excellent review), have a long and respected history of the description of the normal neonatal EEG and its deviation in pathologic states. Accordingly, this history persists in the use of les moˆts justes in its description.

96

Fundamentals: Neonatal Polygraphy

Figure 6-1. The first appearance, maximal expression, and resolution of specific patterns are represented by the gray polygons within each vertical column.

Clinical Pearls 1. Neonatal polygraphy uses a subset of scalp electrodes, EMG, respiratory, cardiac, and eye movement monitoring and is performed for at least 1 hour. 2. ECA ¼ estimated gestational age þ postpartum age. Findings in neonatal EEG are keyed to ECA. 3. The three basic activity states of neonates are wakefulness, active sleep, and quiet sleep. 4. Background EEG activities of neonates are classified as discontinuous or continuous. 5. A lag of expected EEG changes for a given ECA is termed dysmaturity and is one sign of encephalopathy. REFERENCES 1. Guideline 2: Minimum technical requirements for performing pediatric electroencephalography. J Clin Neurophysiol 1994; 3:7–11. 2. Lamblin MD, d’Allest AM, Andre´ M, et al: EEG in premature and full-term infants: Developmental features and glossary. Neurophysiol Clin 1999; 29:1232–1219. 3. Werner SS, Stockard JE, Bickford RG: Atlas of Neonatal Electroencephalography. New York, Raven Press, 1977.

Fundamentals: Neonatal Polygraphy

97

PATIENT 23 Premature infant with intraventricular hemorrhage

An EEG is requested to determine possible ongoing motor activity thought to be seizures in an infant at day 25 postpartum who has an estimated gestational age of 23 weeks. An intraventricular hemorrhage is found. The patient is on no sedative medications. The technologist notes that the patient is asleep and still during this sample. The child is breathing spontaneously on a ventilator. Eye leads are omitted because of nursing staff request. The same sample is shown with a 3-second page (paper speed 10 mm/s) and a 10-second page (30 mm/s). Questions: What is the estimated conceptional age? Name the background activity of this premature infant and the findings at the arrows (shown at a reduced sensitivity in inset). Are these findings normal for this age?

98

Fundamentals: Neonatal Polygraphy

Answers: ECA ¼ 23 weeks þ 25 days *(1/7 weeks/day) ¼ 26 4/7 weeks. The background activity is TD, and the findings at the arrows are temporal theta bursts. Both findings indicate developmentally appropriate background activity for ECA. Discussion: Interpretation of neonatal EEG requires several steps: 1. Calculation of ECA 2. Observation of activity states 3. Recognition of corresponding background activities 4. Identification of physiologic activities upon background activity 5. Determination whether activity states, background activities, and physiologic findings match the stated ECA 6. Identification of pathologic patterns that occur in addition to or instead of expected patterns. Neonates at ECA 35 cps. As a consequence, transients from muscle activity are generally very sharp (35 cps ¼ 1/35 seconds/

cycle ¼ 29 msec/cycle). In this example, prominent, sharp transients at F7 precede apparent slow wave afterpotentials. Lateralis muscle spicules are best seen in anterior temporal leads (F7 and F8). The short duration of the transients and the characteristic slow wave artifacts confirm that the sharp and slow complexes are eye movement artifacts rather than anterior temporal sharp waves.

Clinical Pearl Lateralis muscle spicules are short duration sharp potentials stemming from lateralis muscle artifact during lateral eye movements.

Focal Sharp Transients and Localization-Related Epilepsy

137

PATIENT 39 A 32-month-old boy with generalized seizures

An EEG is requested to evaluate recent onset of generalized convulsions in a 32-month-old boy with recently discovered, poorly characterized amino acid abnormalities. The patient is taking phenobarbital. A previous EEG shows slowing of the waking alpha rhythm and persistent movement artifact from an uncooperative patient, but no IEDs are seen. The current recording is performed after sleep deprivation, with the patient drowsy. The sample is shown in both traverse bipolar and referential montages. Supplementary electrodes at C1 and C2 are placed. Question: Identify the transients (arrows, a) and their clinical significance.

138

Focal Sharp Transients and Localization-Related Epilepsy

Answer: Frequent spikes are present with maximum negativity at the central vertex with potential field asymmetrically shifted to the right. One spike is located at the right central region (a). These are potentially epileptogenic. Discussion: Vertex sharp transients of sleep may be difficult to distinguish from spikes that appear at the central vertex. To illustrate, seizures occur in at least 68% of children with IEDs from temporal, frontal, or occipital foci. Seizures, however, occur in only 38% of children with central IEDs. One possibility is that central spikes are less predictive of epilepsy; another is that some children may have vertex sharp transients of sleep that resemble IEDs. Factors that distinguish vertex sharp transients of sleep from vertex spikes are: Location. Vertex sharp transients appear synchronously and bihemispherically with maximum potential at the vertex. The potential field of vertex spikes, in contrast, is usually asymmetric. A restricted potential field is also a feature of most midline spikes, whereas most vertex sharp transients are broadly distributed.

Supplementary electrodes, C1 and C2 in this example (C1 lies 10% of the coronal distance from midline, or one-half way between Cz and C3), aid in determining a possible asymmetry in the potential field. In this case, the referential montage shows that the amplitude of the negative potential drops quickly from the vertex, but higher amplitudes are present across the right coronal midline. Morphology. Another feature that identifies these transients as IEDs includes the classic morphology of an epileptiform negativity with a slow afterpotential, in contrast to vertex sharp transients that are broadly based and V-shaped. Reactivity/state. Vertex sharp transients of sleep are limited to sleep, whereas central spikes may appear during various states.

Clinical Pearl Spikes that appear at the midline must be carefully distinguished from normal vertex sharp transients of sleep. Asymmetry, restriction of field, and slow afterpotentials distinguish vertex spikes.

REFERENCE 1. Kellaway P: The incidence, significance, and natural history of spike foci in children. In Henry CE (eds): Current Clinical Neurophysiology Update on EEG and Evoked Potential. New York, Elsevier/New Holland, 1981, pp 151–175.

Focal Sharp Transients and Localization-Related Epilepsy

139

PATIENT 40 A 4-year-old girl with spells and cerebral palsy

An EEG is requested to help evaluate possible seizures that consist of head extension, flexion of all four extremities, and cyanosis. The 4-year-old girl has spastic cerebral palsy. The patient is taking baclofen. The patient is asleep for much of the recording. Question: What is the significance of the train of central transients?

140

Focal Sharp Transients and Localization-Related Epilepsy

Answer: Spikes appear with maximum negativity at Cz. Several are equipotential across electrodes Cz and C4, demonstrating a shift of field to the right. Slow afterpotentials are atypical, consisting of a high-amplitude, smoothly contoured positive component and a possible negativepotential slow wave distorted by ongoing delta-theta activities. These are potentially epileptogenic— perhaps. Discussion: Distinction among IEDs and other sharp transients is one of the main challenges of EEG interpretation. This example is particularly difficult. Like pathologic vertex spikes, these spikes and slow waves have a potential field shifted to the right. The typical V-shape of vertex sharp transients of sleep is not present. On the other hand, the transients lack the morphology of a typical negative potential spike

and slow wave complex. Spikes are extremely brief in duration. The slow waves, in this case, are surface-negative slow potentials that follow prominent V-shaped surface positive waves. This was the only burst of such discharges in the recording, thus comparison with more typical vertex sharp transients was not possible.

Clinical Pearls 1. The vertex sharp waves of children can be high amplitude and occur in brief trains that may be difficult to distinguish from IEDs. 2. Accurate reporting of findings and conservative interpretation allow referring physicians to draw their own conclusions in the case of ambiguous findings.

Focal Sharp Transients and Localization-Related Epilepsy

141

PATIENT 41 A 30-year-old woman with major depression with psychotic features

An EEG is requested to evaluate possible encephalopathy in a 30-year-old woman with major depression with psychotic features. Medications are haloperidol and mirtazapine. The recording is made with the patient asleep. Question: Identify the transients (arrows).

142

Focal Sharp Transients and Localization-Related Epilepsy

Answer: Vertex sharp transients of sleep appear widely distributed among all channels, as do sleep spindles, because of a contaminated reference. Discussion: Montage selection is the prerogative of the EEG technologist. Selection of referential montages is particularly tricky because of possible reference contamination, situations in which the potential of interest has an electrical field that involves the common reference. It can lead to nonsensical phase reversals in referential montages, and, at best, distorts familiar activities into unfamiliar ones. In the current case, selection of a Cz reference during light sleep guarantees that midline findings of sleep—vertex sharp transients, K complexes, and sleep spindles—will appear in all channels in an exaggerated fashion that falsely represents the true distribution. Much investigation in the early days of EEG was devoted to the development of the truly uninvolved, noncerebral reference—a kind of Holy Grail. The ear electrode, the original candidate, unfortunately, is involved in many temporal

potentials. In fact, many examples in this book make the use of a paired ear channel that, because of the long interelectrode distance, amplifies potentials extending from temporal regions and acts as a ‘‘signal flag’’ that points up or down and indicates a possible left or right IED. The paired ear channel, however, like other noncerebral sites at the nose or anterior neck, is problematic because of the amplification of EKG artifact. The patient’s state and the location of suspicious potentials are the best guides to the wise selection of reference electrodes. Left temporal abnormalities are best viewed using a reference to the right ear, and vice versa for the other side. Midline references are helpful for bitemporal abnormalities, as long as the state during which they appear is not light sleep. With digital EEG, selection of montages can be done on the fly, so the reviewer is not locked into any one view, a distinct advantage from traditional paper-based systems.

Clinical Pearls 1. Selection of the best referential montage must take into account state and the location of abnormalities. 2. Reference electrodes should be as far away as possible from the location of the putative abnormality.

Focal Sharp Transients and Localization-Related Epilepsy

143

GENERALIZED DISCHARGES AND GENERALIZED EPILEPSY 8

PATIENT 42 A 5-year-old girl with staring spells

A 5-year-old girl has recent onset of numerous brief spells of inattention or staring noted at school and at home. Occasionally, she has limited eye fluttering with spells, but no other automatisms are noted. Her developmental history is normal; in fact, inattention was originally ascribed to boredom in this child, who was exceeding the performance of her classmates until spells became interruptive. The child is awake, hyperventilating, and on no medications. Question:

What is the generalized burst that interrupted hyperventilation?

145

Answer: The recording shows a burst of generalized, rhythmic, high-amplitude spike-wave discharges with a frequency of 3 cps. The episode is provoked by hyperventilation, which, in turn, is interrupted by the event. Diffusely distributed, semirhythmic, medium-amplitude delta activities before the event are also present, representing hyperventilation-induced build-up. This tracing is diagnostic for a generalized, typical absence seizure and is most consistent with childhood absence epilepsy (CAE). Discussion: CAE (petit mal, pyknolepsy) is an idiopathic generalized epilepsy marked by absence seizures. Age of onset ranges from 4–8 years, with a peak at 5–6 years, corresponding to kindergarten in the United States. Patients with CAE experience absence seizures, episodes of behavioral arrest, unresponsiveness, and staring that are sometimes perceived as inattention or daydreaming. Because the seizures are typically brief (less than 15 seconds in about 75% of patients), a history of the spells’ resolution with intervention—‘‘shaking them out of a spell’’—is often unreliable. Mild automatisms, such as eyelid fluttering or blinking, limited orofacial automatisms, or mild alterations in extensor tone, can be present. Absence seizures may occur at a daily frequency too numerous to count, a historical point that is helpful in their distinction from complex partial seizures that occur at a much less frequent rate. Patients usually have no clear antecedent CNS injuries and usually have normal, even supernormal, cognitive abilities. Clinical exam is normal, but hyperventilation can provoke absence seizures in more than half of patients with CAE, a helpful office technique. More than 75% of patients with CAE enjoy total remission of seizures by age 14. Poor prognostic factors for seizure remission include cognitive impairment, susceptibility to hyperventilation, myoclonus with absence, and later age of onset. Patients with continuing absence seizures usually develop generalized tonic-clonic seizures or myoclonic seizures. The EEG is often diagnostic in CAE. Frequency/amplitude. The characteristic finding is rhythmic, 3 cps spike-wave bursts. Amplitudes range from 300–600 mV.

Location. Bursts are generalized, synchronous, and symmetric, usually with anterior dominance in amplitude. Morphology. Rhythmic spike-slow wave complexes, the classic ‘‘dart and dome,’’ are difficult to mistake for other patterns. Longer runs usually demonstrate that frequency, amplitude, and morphology evolve. Initially, the rate of spike-waves ranges between 3 and 3.5 cps, and spikes may attain higher voltages than the following wave. Initial amplitudes are at their highest. At the end of bursts, frequencies may drop to 2.5–3 cps, spikes may fade in amplitude or even disappear, and waves may drop slightly in amplitude. No postictal slowing is present. Activation. Hyperventilation induces spikewave bursts in about 80% of patients with CAE. One characteristic of 3 cps spike-wave is that bursts during non-REM sleep usually become shorter, less regular, and can feature multiple spikes. Viewing discharges only during sleep may lead to a misdiagnosis of atypical spike-wave discharges. Findings that may be confused with 3 cps spike-wave, hypersynchrony of sleep, atypical spike-wave discharges, rhythmic delta activity, and fast multispike-wave discharges will be discussed in separate sections. Obvious clinical changes, such as behavioral arrest seen in the interruption of hyperventilation described previously, may not be present. Confirmation of transient impairment may require response testing. Occipital intermittent rhythmic delta activity (OIRDA) is a frequent finding as well in patients with CAE.

Clinical Pearls 1. Typical 3 cps spike-wave bursts are the key EEG finding in CAE. 2. CAE, when defined by typical spike-wave discharges and not present with cognitive impairment or myoclonus, resolves without residual effects in the majority of children.

146

Generalized Discharges and Generalized Epilepsy

REFERENCES 1. Sato S, Dreifuss FE, J Kirby DD, Palesch Y: Long-term follow-up of absence seizures. Neurology 1983; 33:1590–1595. 2. Weir B: The morphology of the spike-wave complex. Electroencephalography Clin Neurophysiol 1965; 19:284–290. 3. Wirrel E, Camfield C, Camfield P, et al: Long-term prognosis of typical childhood absence epilepsy: remission or progression to juvenile myoclinic epilepsy. Neurology 1996; 47:912–918.

Generalized Discharges and Generalized Epilepsy

147

PATIENT 43 A 7-year-old girl with staring spells

A 7-year-old girl presents with 2–4 staring spells a day. She is performing poorly in school this year after having been successful during the previous year. She takes no medications. The recording is performed during hyperventilation. Question: What is the finding accompanying this spell of inattention? The excerpt shows the middle 5 seconds from parasaggital channels (bar). Fp1–F7 F7–T7 T7–P7 P7–O1 Fp1–F3 F3–C3 C3–P3 P3–O1 Fz–Cz Cz–Pz Fp2–F4 F4–C4 C4–P4 P4–O2 Fp2–F8 F8–T8 T8–P8 P8–O2 A1–A2 T1–F7 T2–F8 E1–A1 E2–A2 EKG

100μV

1s

Fp2–F4 F4–C4 C4–P4 P4–O2 100μV

148

1s

Generalized Discharges and Generalized Epilepsy

Answer: Profound hyperventilation-induced build-up is interrupted by transient inattention. Hyperventilation-induced inattention and rhythmic delta activity (delta absence) probably represents a normal, physiologic response to hypocapnia in individuals who are not suspected of having absence epilepsy. Discussion: Delta absence is a term coined to describe patients undergoing evaluation for staring spells who have episodes of high-amplitude, rhythmic delta activity induced during hyperventilation. Response testing demonstrates that subjects have impaired consciousness during what has since been dubbed hyperventilationinduced high-amplitude rhythmic slowing (HIHARS); in other words, build-up. In the original report, anticonvulsant medications successfully treated staring spells, raising the possibility that rhythmic delta activity was a form of an ictal discharge. Subsequent investigators, however, duplicated both impairment of consciousness and HIHARS in nonepileptic children, suggesting that HIHARS and accompanying inattention is a nonepileptic, physiologic response to hypocapnia. Later work showed that clinical symptoms differ in those patients whose absence seizures with 3 cps spike-wave were activated by hyperventilation and those whose inattention were marked by

HIHARS. Those with 3 cps bursts were more likely to have automatisms during bursts, whereas those with HIHARS merely yawned or fidgeted. In this light, HIHARS most likely represents a nonepileptic phenomenon in subjects who are not suspected of having absence seizures. In these patients, HIHARS is a variant of normal hyperventilation-induced build-up. On the other hand, those with possible absence seizures and HIHARS should be investigated further to uncover further evidence of absence seizures, such as another attempt at recording characteristic 3 cps spike-wave bursts. Hyperventilation-induced buildup may be distinguished from 3 cps spike-wave bursts by the former’s lack of spikes. As seen in the excerpt, ongoing past activities occasionally persist through bursts of delta activity, giving the false appearance of spikes that do not maintain a clear, consistent timing relationship with the slow waves.

Clinical Pearl HIHARS is a variant of normal hyperventilation-induced build-up in patients not suspected of absence seizures.

REFERENCES 1. Epstein MA, Duchowny M, Jayakar P, et al: Altered responsiveness during hyperventilation-induced EEG slowing: A nonepileptic phenomenon in normal children. Epilepsia 1994; 35(6):1204–1207. 2. Lee SI, Kirby D: Absence seizure with generalized rhythmic delta activity. Epilepsia 1988; 29(3):262–267. 3. Lum LM, Connolly MB, Wong PK: Hyperventilation-induced high-amplitude rhythmic slowing with altered awareness: A video-EEG comparison with absence seizures. Epilepsia 2002; 43:1372–1378.

Generalized Discharges and Generalized Epilepsy

149

PATIENT 44 A 41-year-old woman with tonic-clonic convulsions after motor vehicle accident

An EEG is requested to evaluate spells that resemble tonic-clonic convulsions in a 41-year-old woman the day after a motor vehicle accident. She is apparently unharmed but presents with recurrent spells and unresponsiveness in the emergency room where she receives intravenous midazolam. The recording is performed several hours after midazolam administration, with the patient awake. The sample is shown in both bipolar and referential montages. Question: What is the significance of sharp transients noted under the bars?

150

Generalized Discharges and Generalized Epilepsy

Answer: Bursts of diffusely distributed rhythmic theta activities occur during restful wakefulness. Small spikes occur in relationship to theta frequency ‘‘slow waves.’’ This pattern is called phantom spike-wave and is a benign epileptiform transient. Discussion: Phantom spike-waves (6-Hz spike-waves) are benign epileptiform discharges that can be mistaken for pathologic spike-wave discharges. Frequency/location. They appear in bursts with a frequency around 6 cps. Bursts are brief, only lasting 1–2 seconds and appear diffusely across the scalp with varying predominant locations among individuals. Morphology. The phantom spike refers to the small spike associated with the theta activity ‘‘slow wave’’ that often comes and goes during bursts. Reactivity. Phantom spike-wave occurs during restful wakefulness and initial drowsiness. They disappear with light sleep. The main points of distinction between phantom spike-waves and IEDs are that phantom

spike-waves fail to evolve temporally or spatially and are obligately linked to specific states of restful wakefulness or the transition to light sleep. Although older literature suggests that phantom spike-wave is a possible indicator of neurologic or social pathology, phantom spike-wave has no significant clinical correlate and is considered a benign epileptiform transient. In this case, the patient was admitted to an inpatient EEG monitoring unit; continuous video-EEG captured several spells that were not accompanied by ictal discharges. In the situation in which spells are acute and frequent, prompt continuous video-EEG may stave off unnecessary treatment with anticonvulsant medications.

Clinical Pearl Phantom spike-wave is a benign finding of diffusely distributed bursts of theta activities, sometimes occurring with small spike discharges.

Generalized Discharges and Generalized Epilepsy

151

PATIENT 45 A 29-year-old man with medically intractable generalized seizures

A 29-year-old man has a 10-year history of seizures consisting of abrupt loss of consciousness and staring, followed by brief, bilateral tonic posturing of the upper extremities. Frequent, generalized tonic-clonic seizures follow initial symptoms. Most seizures occur nocturnally in clusters. Current medications are lamotrigine and topiramate. The recording is made with the patient asleep. Question: Does the finding here of diffuse spike-wave discharges support a diagnosis of generalized epilepsy?

152

Generalized Discharges and Generalized Epilepsy

Answer: Focal spikes and slowing at F4 (arrow a) before the spike-wave burst, along with initial slow waves or spikes at F4 (arrows b) that precede the burst of diffusely distributed spike-wave complexes (arrows c) suggest a focal epileptic lesion that spreads rapidly and bilaterally. Discussion: Secondary bilateral synchrony designates an EEG finding of generalized and synchronous epileptiform discharges that arise not from generalized epilepsy but from a focal epileptic lesion. Secondary bilateral synchrony should be suspected when generalized spike-wave discharges occur in special relationship to focal findings. Focal IEDs distinct in morphology and recurrent in location are present in addition to generalized discharges. Focal IEDs precede and evolve into secondarily generalized discharges. Generalized spike-wave bursts in secondary bilateral synchrony usually occur in frequencies slower than typical 3 cps spike-wave bursts. Such foci are usually located in the frontal lobes. Frontal lobe seizures may occur briefly and frequently without aura, may be accompanied by subtle asymmetric motor posturing, and may not be followed by postictal confusion. On the other hand, other frontal lobe seizures present with complex, bizarre motor automatisms or focal clonic motor activity. Frontal lobe seizures also tend to occur during sleep and to secondarily generalize. Difficulties arise in the diagnosis between frontal lobe epilepsy with secondary bilateral synchrony and primary generalized epilepsy because both may show interictal focal findings on EEG. In the case of idiopathic generalized epilepsies, about 20% of patients with juvenile myoclonic epilepsy

demonstrate focal spikes. More severe generalized epilepsies can demonstrate focal abnormalities; patients with Lennox-Gastaut syndrome may have complex partial seizures and can demonstrate interictal focal slowing. Several techniques may help in differentiating focal from generalized epilepsies following EEGs that demonstrate possible secondary bilateral synchrony. In patients with generalized or multifocal epilepsies, focal findings tend to be evanescent and usually do not appear on repeat recordings. Examination of generalized bursts of suspected focal origin can be viewed with increased paper speed in an attempt to determine close timing of any initial activity. Bipolar montages may be helpful in minimizing synchronous activity and drawing the eye to focal findings. Ultimately, continuous video-EEG recordings that capture the events in question may be necessary to distinguish between primary and secondary generalized seizures. The assignment of intractable epilepsy patients to either primary generalized or localizationrelated epilepsies aids in guiding therapy. Certain anticonvulsants, carbamazepine and tiagabine, in particular, may exacerbate seizures or provoke status epilepticus in susceptible patients with primary generalized epilepsies. Patients with localization-related epilepsies, even with findings of rapid secondary generalization, may benefit from consideration of epilepsy surgery.

Clinical Pearls 1. Secondary bilateral synchrony designates an EEG finding of bilateral and synchronous epileptiform discharges that arise from a focal epileptic lesion and rapidly generalize. 2. Criteria for secondary bilateral synchrony include that focal IEDs appear temporally and morphologically separate from generalized discharges, and immediately precede generalized discharges. REFERENCES 1. Murthy JM, Rao CM, Meena AK: Clinical observations of juvenile myoclonic epilepsy in 131 patients: A study in South India. Seizure 1998; 7(1):43–47. 2. Perucca E, Gram L, Avanzini G, Dulac O: Antiepileptic drugs as a cause of worsening seizures. Epilepsia 1998; 39:5–17.

Generalized Discharges and Generalized Epilepsy

153

PATIENT 46 A 12-year-old girl with light-provoked seizures and morning myoclonic seizures

A 12-year-old girl has a generalized tonic-clonic seizure while watching cartoons on TV the morning after a late bedtime and sleepover at a friend’s house. She notes habitual and frequent jerks of the upper extremities, clustered in the morning during breakfast and her morning shower. Her mother also has morning myoclonus. The EEG was recorded with the patient awake. Photic stimulation, a rapidly flashing strobe light, is designated in the ‘‘photic’’ channel. Question: What are the findings and the epileptic syndrome that is most likely associated with it?

154

Generalized Discharges and Generalized Epilepsy

Answer: The recording shows a burst of generalized, polyspike wave discharges occurring during and outlasting photic stimulation, defining a photoparoxysmal response. The clinical and electrographic findings are consistent with juvenile myoclonic epilepsy (JME). Discussion: JME (epilepsy of Janz) is an idiopathic, generalized epilepsy marked by myoclonic seizures. Age of onset ranges from 12–19 years, with peak onset between 15 and 16 years. Intelligence is normal. Family history is strongly positive. Family members may experience only minor symptoms, such as morning myoclonus, often identified as ‘‘normal’’ to those who have no reason to think otherwise. Patients with JME experience various generalized seizures: myoclonic seizures (typically clustered in the morning near awakening), generalized tonic clonic seizures, and absence seizures. Seizures in JME are often susceptible to flashing lights or sleep deprivation. JME responds readily to treatment and is associated with excellent cognitive and functional outcome. Remission is rare. Most patients require lifelong treatment.

Generalized polyspike-wave discharges (fast spike-wave, multiple spike-wave) are the typical IEDs seen in JME. Sometimes distinguishing interictal from ictal discharges is difficult, because accompanying jerks may be subtle in some patients. Frequency/location. The generalized, often anteriorly dominant, bursts of polyspike-wave discharges typically occur at a much faster rate than typical spike-wave seizures in CAE. Spikes

occur in ~5 cps or faster within bursts, tending to slow within single bursts. Morphology. As shown in the excerpt, the morphology of individual discharges can vary within the same run. Bursts of spike-wave discharges can appear intermixed with bursts of spikes, typically in rhythmic runs in the alpha or theta range, followed by prominent negative-potential slow waves. Activation. As demonstrated in the present case, photic stimulation provokes a burst of generalized, polyspike-wave discharges—photoparoxysmal responses—that outlast photic stimuli. Note that the technologist has terminated each block of stimuli prematurely (usually blocks are 10 seconds long in our laboratory) for fear of provoking generalized tonic-clonic seizures. Because treatment with anticonvulsant medications can decrease the incidence of abnormal discharges in the case of idiopathic generalized epilepsies, reasonable attempts at a confirmatory EEG before treatment is recommended. Finally, because seizures in JME are more likely to occur in the morning or after awakening, scheduling recordings in the morning after sleep deprivation may further improve sensitivity. Photosensitivity provokes photoparoxysmal responses in about 50% of women with JME and in about 25% of men. Positive responses are much higher in those who do not experience generalized convulsions (100% women, 50% men). JME must be differentiated from progressive myoclonic epilepsies, an assortment of diseases associated with neurologic deterioration, myoclonus epilepsy, and poor outcome. As opposed to many of these diseases, the background EEG in JME is normal.

Clinical Pearls 1. Juvenile myoclonic epilepsy is a syndrome of adolescent onset of generalized seizures, usually myoclonic and generalized tonic clonic, associated with excellent response to treatment, cognitive sparing, and lack of remission. 2. The IED seen in JME is the generalized polyspike-wave discharge. The finding occurs spontaneously and during photic stimulation. Polyspike-wave discharges during photic stimulation are called photoparoxysmal responses.

Generalized Discharges and Generalized Epilepsy

155

REFERENCES 1. Janz D: The idiopathic generalized epilepsies of adolescence with childhood and juvenile age of onset. Epilepsia 1997; 38:4–11. 2. Janz D, Durner M, Beck-Mannagetta G, Pantazis G: Family studies on the genetics of juvenile myoclonic epilepsy (epilepsy with impulsive petit mal). In Beck-Mannagetta G, Anderson VE, Doose H, Janz D (eds): Genetics of the Epilepsies. Berlin, Springer, 1989, pp 43–52. 3. Wolf P, Gooses R: Relation of photosensitive epilepsy to epileptic syndromes. J Neurol Neurosurg Psychiatry 1986; 49:1386–1391.

156

Generalized Discharges and Generalized Epilepsy

PATIENT 47 A 14-year-old girl with photic discomfort

A 14-year-old girl has visual discomfort provoked by working on computers. Family history is significant for epilepsy in an older sister with JME. The recording is performed with the patient awake and on no medications. Question:

What does the response during photic stimulation indicate?

Generalized Discharges and Generalized Epilepsy

157

Answer: The burst of generalized polyspike-wave discharges does not outlast photic stimulation. Self-limited photoparoxysmal responses have unclear clinical significance. Discussion: The most significant abnormal response during photic stimulation is the photoparoxysmal response, a burst of generalized polyspike-wave discharges provoked by photic stimulation. Photoparoxysmal responses are traditionally divided into two groups, those that continue after photic stimulation ends (‘‘prolonged’’), and those that cease after photic stimulation (‘‘self-limited’’). Prolonged photoparoxysmal responses are thought to have a significant association with idiopathic generalized epilepsy. JME has the highest incidence of association with photoparoxysmal responses. It is one of the few EEG findings that are not equally split between the sexes; women have a higher incidence of photosensitivity. In contrast, self-limited photoparoxysmal responses may occur in nonepileptic subjects or

asymptomatic family members. No data suggest just how ‘‘prolonged’’ is significant. Many recommend that one waveform, that is, one full polyspike-wave, after cessation of stimulation is abnormally prolonged. Other studies show a significant association between photoparoxysmal responses and idiopathic, generalized epilepsies regardless of the self-limited or prolonged nature of the discharge. A conservative approach is to interpret selflimited photoparoxysmal responses as those with unclear clinical significance, noting that familial traits may allow their expression. The finding must be placed in context with the symptoms of the patient. In this particular patient, photoparoxysmal responses remained self-limited to photic blocks. No spontaneous IEDs were observed. Symptoms gradually attenuated over the next year in follow-up.

Clinical Pearls 1. Photoparoxysmal responses of generalized polyspike-wave discharges that last beyond blocks of photic stimuli are seen in JME and in other idiopathic, generalized epilepsies. 2. Self-limited photoparoxysmal responses that do not outlast stimuli have unclear clinical significance and can be an indicator of a familial trait. REFERENCES 1. Puglia J, Brenner R, Soso M: Relationship between prolonged and self-limited photoparoxysmal responses and seizure incidence: Study and review. J Clin Neurophysiol 1992; 9(1):137–144. 2. Reilly EW, Peters JF: Relationship of some varieties of electroencephalographic photosensitivity to clinical convulsive disorders. Neurology 1977; 23:1045–1057. 3. Wolf P, Gooses R: Relation of photosensitive epilepsy to epileptic syndromes. J Neurol Neurosurg Psychiatry 1986; 49:1386–1391.

158

Generalized Discharges and Generalized Epilepsy

PATIENT 48 A 12-week-old female infant with spells of bilateral limb extension and trunk flexion

A 12-week-old female infant is delivered at term via an uncomplicated pregnancy. She develops spells of bilateral limb extension and trunk flexion after first having spells of rightward head deviation and right arm extension. Coinciding with spells is the mother’s report of decreased activity and feeding. The patient takes no medications. The recording is made with the patient awake and asleep. Clinical sleep is shown in the 20second sample. Question:

What is the epileptic syndrome?

Generalized Discharges and Generalized Epilepsy

159

Answer: The EEG shows no age-appropriate activities of sleep; instead, recurrent high-amplitude bursts of activity are interrupted by periods of relatively suppressed activity. Suppression-burst with episodes of infantile spasms are consistent with Ohtahara syndrome. Discussion: Epileptic encephalopathy defines a syndrome of decline in function or regression in development, epileptic seizures, and stereotypic patterns of interictal activity on EEG. Characteristics of each element of this triad depend on age and may evolve from one to another. Epileptic encephalopathies should be considered as age-specific pathologic responses to various brain injuries. Epileptic encephalopathies can be classified as idiopathic, symptomatic, or cryptogenic epilepsies, depending upon the specific etiology. Ohtahara syndrome, also known as early infantile epileptic encephalopathy (EIEE), occurs within the first 3 months of life and is marked by developmental regression, infantile spasms or partial seizures, and a suppression-burst pattern on the interictal EEG. Ohtahara syndrome should be distinguished from early neonatal myoclonic encephalopathy (ENME). Both syndromes appear in the neonatal period, and both are marked by suppression-

burst. ENME presents with massive myoclonus, rather than infantile spasms. ENME is usually caused by progressive and often devastating etiologies stemming from inborn errors of metabolism. Glycine encephalopathy is a classic cause of ENME presenting with suppression-burst on EEG. Ohtahara syndrome, on the other hand, is often caused by cerebral dysgenesis and can have idiopathic causes, implying a better outcome. Response to treatment (ACTH in the United States, often vigabatrin outside the United States) is also more variable than in ENME. Progression to later epileptic encephalopathies is common. Follow-up EEGs are helpful in determining the course of therapy, and resolution of suppressionburst early in the course is thought to be a good prognostic sign. This patient was treated with ACTH injections. Infantile spasms resolved, but by age 6 months the child was not progressing developmentally. No specific etiology was uncovered.

Clinical Pearls 1. Epileptic encephalopathies are syndromes defined by the triad of clinical regression, seizure type, and interictal EEG pattern. 2. Ohtahara syndrome is diagnosed by neonatal regression, infantile spasms, and suppression-burst pattern on interictal EEG. REFERENCES 1. Tharp BR: Neonatal seizures and syndromes. Epilepsia 2002; 43:S2–S10. 2. Yamatogi Y, Ohtahara S: Early-infantile epileptic encephalopathy with suppression-bursts: Ohtahara syndrome; its overview referring to our 16 cases. Brain Dev 2002; 24:13–23.

160

Generalized Discharges and Generalized Epilepsy

PATIENT 49 An 8-month-old male infant with congenital abnormalities and spells of bilateral limb flexion

An 8-month-old male infant has midline congenital abnormalities and presents with developmental delay and spells of bilateral limb and trunk flexion that cluster upon awakening from naps. The patient is taking no medications. The recording is made with the patient awake after having awoken from a nap. An episode of trunk and bilateral limb flexion (‘‘body jerk’’) is noted by the technologist. Question:

What are the EEG findings and the epileptic syndrome?

Generalized Discharges and Generalized Epilepsy

161

Answer: Hypsarrhythmia. Attenuation that corresponds to an infantile spasm is an electrodecremental seizure. These findings are consistent with West syndrome. Discussion: The triad of developmental delay, hypsarrhythmia, and infantile spasms defines West syndrome. West syndrome is an epileptic encephalopathy that affects infants and young children from ages 3 months to 2 years, with peak onset at ages 8–9 months. Infantile spasms consist of abrupt and brief flexion of the upper limbs and trunk, sometimes along with the legs. The rapid flexions lead to the obsolete terms ‘‘clasp-knife’’ or ‘‘salaam’’ seizures. Occasionally, infantile spasms occur as extensor events. Seizures often cluster at transitions from sleep to wakefulness. As in Ohtahara syndrome, various etiologies are associated with West syndrome, although many are eventually classified as cryptogenic. Tuberous sclerosis is an important cause of symptomatic West syndrome. Although EEG changes are diffuse, functional neuroimaging may reveal a lesion amenable to epilepsy surgery. Outcome in West syndrome is determined by the underlying cause, although control of the seizures and hypsarrhythmia—usually with ACTH in the United States and with vigabatrin elsewhere—is associated with better cognitive outcome. About a third of patients progress to the next epileptic encephalopathy, Lennox-Gastaut syndrome. Hypsarrhythmia means ‘‘mountainous activities.’’ Normal background activities are absent;

instead, continuous arrhythmic, high-amplitude, asynchronous delta activities with independent, multifocal spike discharges are present. Many interpreters specify a minimum amplitude of 200 mV to qualify as hypsarrhythmia. Sleep may cause continuous hypsarrhythmia to fragment, allowing periods of lower-amplitude semirhythmic delta and theta activities to emerge. When delta and theta background activities are present during both sleep and wakefulness, interspersed between runs of hypsarrhythmia, the pattern is traditionally called modified hypsarrhythmia. Despite the apparent ‘‘improvement’’ by the emergence of background activity, modified hypsarrhythmia has the same clinical import as continuous hypsarrhythmia. An abrupt attenuation that follows a high-amplitude sharp-and-slow wave complex, electrodecremental seizure, is the usual marker of an infantile spasm. This patient was eventually found to have multiple heterotopias, polymicrogyria, and dysmyelination on neuroimaging. ACTH injections resolved hypsarrhythmia and spasms, but the child did not survive because of complications due to other malformations.

Clinical Pearls 1. West syndrome is a triad of infantile developmental regression or delay, infantile spasms, and hypsarrhythmia. 2. Hypsarrhythmia is defined as continuous, high-amplitude, arrhythmic, asynchronous delta activities with interspersed independent, multiple spikes. REFERENCES 1. Asano E, Chugani DC, Juhasz C, Muzik O: Surgical treatment of West syndrome. Brain Dev 2001; 23:668–676. 2. Dulac O: Epileptic encephalopathy. Epilepsia 2001; 43:S23–S26. 3. Gibbs FA, Gibbs EL: Atlas of Electroencephalography. Cambridge, MA, Addison-Wesley, 1952.

162

Generalized Discharges and Generalized Epilepsy

PATIENT 50 A 3-year-old girl with developmental delay and ‘‘drop attacks’’

A 3-year-old girl with developmental delay is referred for evaluation of recent onset of drop attacks, abrupt loss of tone causing her to fall forward. She is taking valproate. The recording is made with the patient awake. The G2 input is the averaged ear reference (A1 þ A2). Question:

What are the EEG findings and the epileptic syndrome?

Generalized Discharges and Generalized Epilepsy

163

Answer:

Slow spike-wave consistent with Lennox-Gastaut syndrome.

Discussion: The triad of developmental delay, astatic seizures, and atypical spike-wave bursts defines Lennox-Gastaut syndrome (LGS). LGS is an epileptic encephalopathy that affects children from ages 2–18. As in other epileptic encephalopathies, LGS can result from idiopathic, symptomatic, or cryptogenic causes. Many patients with LGS had a previous epileptic encephalopathy, such as West syndrome. Authorities emphasize different seizures in Lennox-Gastaut. Astatic seizures (drop attacks) certainly can be the most difficult to treat and the most injurious. Atypical absences, featuring more myoclonus and automatisms than typical absence seizures, frequently occur, as well as tonic and tonic-clonic seizures. Most now refer to the various mixed, generalized seizures that often remain refractory to treatment. Slow spike-wave, bursts of rhythmic generalized spike-wave discharges, must be differentiated

from typical 3 cps spike-wave bursts. Slow spike-wave bursts have a frequency of 2–2.5 cps, as opposed to 3–3.5 cps of typical spike-wave. Often the ‘‘spikes’’ of slow spike-wave bursts are broader than 70 milliseconds. Bursts of slow spike-wave may be also less rhythmically regular than typical spike-wave. Slow spike-wave bursts in LGS have no clinical accompaniment, whereas typical spike-wave bursts usually denote clinical and electrographic seizures. These differences lead to the syndrome’s original name of petit mal variant. Other abnormalities in the EEG of LGS include abnormally slow background activities of wakefulness and independent, multifocal spikes. Sleep recordings may contain bursts of low-amplitude, generalized, rhythmic alpha-beta activities, similar to electrodecremental seizures, that usually mark occurrence of tonic seizures.

Clinical Pearls 1. Lennox-Gastaut syndrome is a triad of childhood developmental regression or delay, mixed generalized seizures, and slow spike-wave. 2. A slow spike-wave is an interictal pattern in Lennox-Gastaut syndrome that, unlike typical 3 cps spike-wave, occurs at a frequency of 2–2.5 cps. REFERENCES 1. Dulac O: Epileptic encephalopathy. Epilepsia 2001; 43:S23–S26. 2. Gibbs FA, Gibbs EL: Atlas of Electroencephalography. Cambridge, MA, Addison-Wesley, 1952.

164

Generalized Discharges and Generalized Epilepsy

PATIENT 51 A 4-year-old boy with progressive neurologic deterioration and myoclonic epilepsy

Ongoing seizure activity is questioned in a 4-year-old boy with the subacute onset of myoclonic seizures and cognitive deterioration. He is admitted for nearly continuous generalized and migratory myoclonic seizures. Chronic anticonvulsants include lamotrigine and zonisamide after the patient failed trials of valproate and topiramate. The recording is performed with the patient awake but unable to follow commands. He is given intravenous lorazepam 3 hours before the recording. Questions:

What is the finding? With what epilepsy syndrome is it associated?

Generalized Discharges and Generalized Epilepsy

165

Answers: The finding is occipital spikes evoked by photic stimulation (high-amplitude photic driving responses). Giant visual evoked potentials and somatosensory evoked potentials may be seen in certain progressive myoclonus epilepsies. Discussion: Progressive myoclonus epilepsies (PME) encompass a group of symptomatic, generalized epilepsies denoted by cognitive deterioration to the point of dementia, neurologic deficits referable to multiple systems, and myoclonic seizures. Myoclonic seizures can be massive, generalized jerking, or migratory, segmental myoclonus. Partial seizures, especially from the occipital lobe, are also prevalent. Diseases causing PME include UnverrichtLundborg disease (Baltic myoclonus), sialidosis (cherry-red spot myoclonus), Gaucher’s disease (glucocerebroside beta-glucosidase deficiency), mitochondrial encephalopathy with ragged red fibers (MERRF), Lafora’s disease, and neuronal ceroid lipofuscinosis. The background EEG in patients with PME is marked by replacement of normal waking and sleeping activities with abnormally slow and disorganized theta and delta activities. Multifocal or generalized spikes, polyspikes, and atypical

spike-wave discharges are often present. Occipital spikes can be present. Although most patients with PME have unremarkable evoked potentials, some patients, especially those with neuronal ceroid lipofuscinosis, have abnormally high amplitude evoked potentials. High-amplitude somatosensory evoked potentials, so-called giant SSEPs, are thought to reflect a state of heightened cortical excitability. Analogous to giant SSEPs, high-amplitude pattern reversal visual evoked potentials can also be present. In this case, the patient has the late infantile subtype of neuronal ceroid lipofuscinosis (JanskyBielschowsky disease). Prominent occipital spikes, reminiscent of giant visual-evoked potentials, were recorded only during slow-frequency photic stimulation. Whereas photoparoxysmal response are most typically evoked at stimulation frequencies of 10–20 Hz, high-amplitude driving responses in susceptible individuals with PME are best seen at slower driving frequencies between 2 and 5 Hz.

Clinical Pearls 1. Progressive myoclonic epilepsies are a group of diseases marked by progressive cognitive and neurologic deterioration and myoclonic epilepsy. 2. The EEG in PME is distinguished by abnormally slow and disorganized background activities, and multifocal spike, polyspike, or atypical spike-wave discharges. 3. High-amplitude (‘‘giant’’) posterior driving responses evoked at slow photic driving frequencies can be seen in neuronal ceroid lipofuscinosis. REFERENCES 1. Pampiglione G, Harden A: So-called neuronal ceroid lipofuscinosis: Neurophysiological studies in 60 children. J Neurol Neurosurg Psychiatry 1977; 40:323–330. 2. Scaioli V, Nardocci N: A pathophysiological study of neuronal ceroid lipofuscinoses in 17 patients: Critical review and methodological proposal. Neurol Sci 2000; 21:S89–S92. 3. Schmitt B, Thun-Hohenstein L, Molinari L, et al: Somatosensory evoked potentials with high cortical amplitudes: Clinical data in 31 children. Neuropediatrics 1994; 25:74–84.

166

Generalized Discharges and Generalized Epilepsy

PATIENT 52 A 7-month-old girl with generalized seizures after anoxic encephalopathy

A 7-month-old girl status post anoxic encephalopathy developed generalized tonic clonic and myoclonic seizures. Medications are phenobarbital and levetiracetam. In an attempt to catch a seizure, a prolonged recording—2 hours total—was performed in the laboratory. Question:

What are the locations of spikes labeled at time a, b, and c? a

b

c

Fp1–F7 F7–T7 T7–P7 P7–O1 Fp1–F3 F3–C3 C3–P3 P3–O1 Fz–Cz Cz–Pz Fp2–F4 F4–C4 C4–P4 P4–O2 Fp2–F8 F8–T8 T8–P8 P8–O2 A1–A2 T1–F7 T2–F8 EKG E1–A1 E2–A2

Generalized Discharges and Generalized Epilepsy

167

Answer:

(a) Left midtemporal, (b) right parietal, and (c) bifrontal regions.

Discussion: IEDs that occur in several different locations with no clear relationship to each other are termed multifocal, independent spikes. Like the patient shown here, multifocal independent spikes usually occur in subjects with static or progressive encephalopathies. Most patients have frequent seizures that are medically intractable, and most are mentally retarded. Because of this

association with severe seizures, some EEGers designate multifocal, independent spikes as ‘‘highly epileptogenic.’’ In this patient with anoxic encephalopathy, background EEG activities of diffuse, unreactive arrhythmic delta activities reflect the severe involvement of the disease’s effect on white matter.

Clinical Pearls 1. Multifocal, independent spikes usually occur in the setting of background slowing of the EEG in patients with static or progressive diffuse or multifocal brain disease. 2. Multifocal, independent spikes are usually considered highly epileptogenic because of their strong association with medically–intractable epileptic seizures.

168

Generalized Discharges and Generalized Epilepsy

OTHER SEIZURE SYNDROMES 9

PATIENT 53 A 28-year-old woman after a single seizure

A 28-year-old previously healthy woman presents after a single event of staring and confusion with mild oral-facial automatisms after an all-night party. She has no epilepsy risk factors, such as febrile convulsions, head trauma, or CNS infection. Family history is unremarkable for epilepsy. Neurologic examination is normal. A previous EEG is normal. She is taking no medications. The EEG is recorded with the patient awake following overnight sleep deprivation. Questions: What is the interpretation of the finding in the left temporal region (arrows a and b)? What diagnosis does this finding support?

169

Answers: A spike (arrow b) localized to the left anterior temporal region and a burst of arrhythmic slowing (arrow a) indicate a physiologic or structural lesion of the left temporal region that is potentially epileptogenic. This finding supports a diagnosis of epileptic seizure, accounting for the patient’s single episode, and implies a risk of further seizures. Discussion: The clinical presentation of a single, new-onset seizure is one of the most common problems facing the electroencephalographer and neurologist. IEDs and epileptic seizures maintain a robust but not absolute correlation. The sensitivity of a single routine EEG in the yield of IEDs in known epilepsy is approximately 50%. Repeating routine studies increases sensitivity but with gradually diminishing returns, eventually attaining 80–85% after the fourth recording. The specificity of IEDs depends on the surveyed population. For example, in adults with transient loss of consciousness from either syncope or seizure, IEDs were specific for seizure in 95%. On the other hand, epileptiform activity appears in 0.4–0.5% of adults and 1.5–3.5% of children without seizures or other neurologic disorders. The clinical import of finding IEDs in asymptomatic individuals is controversial, especially in determining work-related risks. For example, in studies of military aircrew, subjects with IEDs on screening EEG developed epilepsy at an incidence not much different to that of the base population (2%). On the other hand, other studies calculate risks of developing epilepsy in this supernormal population at 25%.

The clinical significance of IEDs in patients with suspected seizure, however, is much clearer. EEG helps establish the type of seizure and epilepsy syndrome, which are important for prognosis and treatment. Seizure etiology and EEG findings are the strongest predictors of seizure recurrence in adults. Meta-analyses across many studies show that, for all adults, risk of recurrent seizures following an initial, unprovoked seizure is  40%. The risk is highest for those with a presumptive symptomatic cause and EEG findings consistent with focal abnormalities: 65%. The variable sensitivity of IEDs in patients, on the other hand, indicates that a lack of IEDs should not be taken as evidence against seizures. In these subjects, repeated study following sleep deprivation and containing sleep should be performed. Recordings performed soon after the event in question are more likely to show IEDs. In the present case, neuroimaging revealed no abnormalities. Based on findings of left temporal spikes and slowing, the primary neurologist recommended treatment with anticonvulsant medication for presumptive localization-related epilepsy and risk of recurrent seizures of >50%. The patient elected to not take anticonvulsant medications and not drive for her state’s proscribed duration. Carbamazepine was started after a second seizure occurred 2 months after this tracing.

Clinical Pearls 1. The sensitivity of IEDs in known epilepsy is about 50% for a single study and about 80–85% for up to four studies. 2. The specificity of IEDs in healthy adults without epilepsy is controversial, but in highly selected individuals, the risk of subsequent seizures appears no greater than the base population. 3. The specificity of IEDs in subjects with suspected seizures is high and depends on the study group. The risk of recurrent seizures in adults with suspected symptomatic causes and accompanied by focal abnormalities on the EEG is 65%. REFERENCES 1. Berg AT, Shinnar S: The risk of seizure recurrence following a first unprovoked seizure: A quantitative review. Neurology 1991; 41(7):965–972. 2. Gotman J, Marciani MG: Electroencephalographic spiking activity, drug levels, and seizure occurrence in epileptic patients. Ann Neurol 1985; 17(6):597–603. 3. Gregory RP, Oates T, Merry RT: Electroencephalogram epileptiform abnormalities in candidates for aircrew training. Electroencephalogr Clin Neurophysiol 1993; 86(1):75–77. 4. Hendriksen IJ, Elderson A: The use of EEG in aircrew selection. Aviat Space Environ Med 2001; 72(11):1025–1033.

170

Other Seizure Syndromes

5. Hoefnagels W, Padberg G, Overweg J, et al: Syncope or seizure? The diagnostic value of the EEG and hyperventilation test in transient loss of consciousness. J Neurol Neurosurg Psychiatry 1991; 54(11):953–956. 6. Salinsky M, Kanter R, Dasheiff RM: Effectiveness of multiple EEGs in supporting the diagnosis of epilepsy: An operational curve. Epilepsia 1987; 28:331–334.

Other Seizure Syndromes

171

PATIENT 54 A 6-year-old boy after a single generalized convulsion

A 6-year-old, previously healthy boy has a generalized convulsion while walking with his mother. He has no epilepsy risk factors, such as febrile convulsions, head trauma, or CNS infection. Family history is unremarkable, and neurologic examination is normal. He is taking no medications. The EEG is recorded with the patient awake following overnight sleep deprivation. The sample is reformatted in bipolar longitudinal, referential, and transverse bipolar montages. Questions: Describe the location and polarity of the spikes. What diagnosis for the patient’s spell do these findings support?

172

Other Seizure Syndromes

Other Seizure Syndromes

173

Answers: High-amplitude spikes (125–150 mV) have a morphology resembling Rolandic spikes. In the longitudinal montage, they appear with negativity at end of chain beyond Fp1 and O1 and O2. Negative phase reversal is apparent at F3, and possible positive phase reversals are apparent at C3 and P3. In the referential montage, maximum amplitudes are apparent at Fp1 and O1/ O2. In the transverse view, negative phase reversals appear both at Fp1 and at O1/O2. The pattern in most consistent with a broad negativity deep within the left hemisphere with a projected, vertically oriented positive dipole. The spikes are epileptogenic but do not fall within any clear epilepsy syndrome. Although supportive of a diagnosis of epileptic seizure, the risk of seizure recurrence is unclear, despite the definitive findings of IEDs. Discussion: The single seizure during childhood presents a different combination of challenges than that during adulthood. First, unlike adult syndromes, many childhood epilepsy syndromes are self-limited in course. Second, the social and economic effects of seizures in children are arguably less harmful than in adults. Third, children may be more susceptible to the cognitive effects of anticonvulsant medications than adults. These factors make some pediatric neurologists more likely to defer treatment with anticonvulsant medications until seizures clearly appear recurrent. In support of this strategy are studies that find that IEDs during childhood have a less strong association with epilepsy. For example, the incidence of IEDs in healthy adults is about 0.4– 0.5%, whereas the incidence is threefold or greater (1.5–3.5%) in healthy children. Nevertheless, children with IEDs on EEG have a twofold relative risk of seizure recurrence after a first seizure than children without IEDs. EEG findings are most helpful when they help establish an epilepsy syndrome. For example, children

with presumed idiopathic and cryptogenic syndromes have a 30–50% rate of recurrence of seizures within 2 years, whereas those with ‘‘remote symptomatic causes’’ (history of brain insult, such as mental retardation or cerebral palsy) have a >50% chance of recurrence. In this case, spikes were present during both wakefulness and sleep. Although the spikes in this case have a morphology reminiscent of Rolandic spikes and a dipole, they are atypical from those seen in benign Rolandic epilepsy because of their basal location and persistence in all states; Rolandic spikes appear in centrotemporal regions preferentially during drowsiness. Similarly, the frontal and leftward electrical fields of the current spikes are atypical for benign occipital lobe epilepsy of childhood. Also, the spikes do not appear to emanate from a clear epileptic lesion. In the current case, therefore, no particular epilepsy syndrome is apparent. The primary neurologist for this patient recommended ‘‘expectant observation’’ rather than treatment with anticonvulsant medications.

Clinical Pearls 1. The incidence of IEDs in healthy children (1.5–3.5%) is higher than that of older age groups (0.5%). 2. The adage ‘‘treat the patient, not the EEG’’ is most wisely followed in children; findings of IEDs in children must be placed in the context of risk and benefit. 3. Viewing selected findings with different montages, feasible with most digital EEG systems, is recommended for accurate localization. REFERENCES 1. Berg AT, Shinnar S: The risk of seizure recurrence following a first unprovoked seizure: A quantitative review. Neurology 1991; 41(7):965–972. 2. Camfield C, Camfield P, Gordon K, Dooley J: Does the number of seizures before treatment influence ease of control or remission of childhood epilepsy? Not if the number is 10 or less. Neurology 1996; 46(1):41–44. 3. Cavazzuti GB, Cappella L, Nalin A: Longitudinal study of epileptiform EEG patterns in normal children. Epilepsia 1980; 21(1):43–45. 4. Eeg-Olofsson O, Petersen I: The development of the EEG in normal children from the age of 1 to 15 years: Paroxysmal activity. Neuropadiatrie 1971; 4:375–404. 5. Hirtz D, Berg A, Bettis D, et al: Practice parameter: Treatment of the child with a first unprovoked seizure: Report of the Quality Standards Subcommittee of the American Academy of Neurology the Practice Committee of the Child Neurology Society. Neurology 2003; 60(2):166–175.

174

Other Seizure Syndromes

PATIENT 55 A 3-year-old boy after a febrile convulsion

An EEG is requested to evaluate possible ongoing seizure in a 3-year-old boy who has recurrence of febrile convulsions first experienced 13 months prior. Unlike previous episodes that occurred with tympanic or rectal temperatures exceeding 38.5 C, rectal temperature during the current episode is 38 C. The patient takes no medications. The recording is performed with the patient intermittently agitated and lethargic. Eye leads are not placed because of the agitated state. Questions: Beyond diffuse slowing inappropriate for waking state, what is the other abnormality shown in this sample? Does the abnormality help in prognosis of epilepsy?

Other Seizure Syndromes

175

Answers: Arrhythmic delta activity is broadly distributed but focal delta activity appears in the left posterior temporal region. Localized abnormalities seen after seizure or postictal slowing are evidence of focal neuronal dysfunction. Although nonepileptiform abnormalities are associated with higher risk of afebrile seizure occurrence, epileptiform abnormalities represent a higher risk. Discussion: Simple febrile convulsions are generalized seizures triggered by fever that is not due to CNS infection. Peak incidence occurs around age 18 months and is unusual past 3 years. Complex or atypical febrile convulsions, in contrast to simple febrile convulsions, are prolonged (>30 minutes) or have partial, rather than generalized, features. In the case of simple febrile convulsions, the EEG offers little diagnostic information. Diffuse, often occipitally dominant postictal slowing may be evident if the recording is performed soon after resolution of the seizure. In the case of atypical febrile convulsions, or those convulsions that occur without a definite fever, the EEG may be more useful. Studies of new-onset seizures in children (not just febrile convulsions) show that the risk of recurrence is highest in those with IEDs on EEG (about two

times those without abnormalities on EEG). The risk of recurrence is only marginally higher than those with normal EEGs if the abnormality is nonepileptiform (1.3 times). However, the risk of recurrence of febrile convulsions appears to be most strongly linked to characteristics of the fever than to seizure type or electrographic findings. An increased risk of recurrent seizures is associated with a short duration of fever and a relatively lower maximum temperature. In the present case, the predominance of arrhythmic delta activity across the left hemisphere was interpreted as abnormal and suggestive of localized nonspecific pathology. Repeat recordings are usually recommended to see if presumed focal abnormalities persist. Later cases will further discuss the appearance and significance of focal slowing.

Clinical Pearls 1. The use of EEG in the evaluation of simple febrile convulsions is unclear. 2. Focal abnormalities or IEDs in atypical febrile convulsions and other new-onset seizures in children are associated with a higher risk of spontaneously recurrent seizures. REFERENCES 1. Berg A, Hauser W, Alemany M, et al: A prospective study of recurrent febrile seizures [see comments]. N Engl J Med 1992; 327(16):1122–1127. 2. Berg AT, Shinnar S: The risk of seizure recurrence following a first unprovoked seizure: A quantitative review. Neurology 1991; 41(7):965–972.

176

Other Seizure Syndromes

PATIENT 56 An 8-year-old girl with acquired aphasia and generalized convulsions

An 8-year-old girl has a 1-year history of progressive behavioral difficulties and, more recently, worsening receptive aphasia. Occasional generalized tonic-clonic seizures started about a year prior. She is treated with valproate. The recording is performed with the patient awake. Questions: What is the location of the spikes in the tracing? What epilepsy syndrome does this likely represent?

Other Seizure Syndromes

177

Answers: Spikes have maximum negativity in posterior temporal and central regions, with a single generalized burst of spike-wave. The clinical and electrographic findings are consistent with Landau-Kleffner syndrome. Discussion: Landau-Kleffner syndrome is a rare syndrome with striking features of childhood onset of acquired aphasia and epileptic seizures or an EEG showing IEDs. Psychiatric abnormalities may be present and, in some cases, resemble autism. Receptive, rather than expressive, aphasia predominates. Generalized seizures tend to be less dramatic and more easily treated than behavioral or language abnormalities. Landau-Kleffner syndrome is usually associated with focal spikes from centroparietotemporal regions. The broader term epileptic aphasia encompasses a range of related disorders that all present with various severities of behavioral abnormalities, language regression, and epilepsy. Electrical status epilepticus of sleep (ESES), sometimes referred to as continuous spike-wave of slow wave sleep, is an EEG finding of spike-wave discharges present during more than 85% of slow wave sleep. Sleep activates IEDs in all of these syndromes, and the waking EEG can be normal. In fact, the

example shown here is unusual in that the characteristic IEDs are evident during wakefulness and sleep. Because of the symptomatic overlap with autism, many centers are asked to perform EEGs as part of the evaluation of autism. If the initial daytime recording is normal and epileptic aphasia is still suspected, an overnight recording as an inpatient or with the use of outpatient ambulatory EEG should be performed to capture an adequate sample of slow wave sleep. Admittedly, the behavioral challenges present in these children sometimes make adequate sampling of sleep difficult. Lately, some report that rare patients with syndromes similar to benign childhood epilepsy with centrotemporal spikes (Rolandic epilepsy) progress to epileptic aphasia. Although centrotemporal spikes can be present in both groups, no reports have yet determined what predictive factors may exist.

Clinical Pearls 1. Landau-Kleffner syndrome consists of psychiatric abnormalities, receptive aphasia, and generalized seizures. 2. IEDs in epileptic aphasia are activated by, and may be confined to, sleep; adequate EEG evaluation of epileptic aphasia must include slow-wave sleep. 3. ESES consists of continuous runs of spike-wave discharges that occupy slowwave sleep and is one of the findings in syndromes of epileptic aphasia. REFERENCES 1. Fejerman N, Caraballo R, Tenembaum SN: Atypical evolutions of benign localization-related epilepsies in children: Are they predictable? Epilepsia 2000; 41:380–390. 2. Landau WM, Kleffner FR: Syndrome of acquired aphasia with convulsive disorder in children. Neurology 1957; 7:523–530. 3. Tassinari CA, Rubboli G, Volpi L, et al: Encephalopathy with electrical status epilepticus during slow sleep or ESES syndrome including the acquired aphasia. Clin Neurophysiol 2000; 111:S94–S102.

178

Other Seizure Syndromes

ICTAL DISCHARGES AND EPILEPTIC SEIZURES 10

PATIENT 57 A 24-year-old man with a right occipital cystic lesion and right mesial temporal lobe epilepsy

A 24-year-old man has medically intractable complex partial seizures with auras. He has dual pathology of a hamartoma in the right occipital lobe and right hippocampal sclerosis. Scalp videoEEG during evaluation for epilepsy surgery discloses independent, bitemporal seizure onsets, with symptoms preceding electrographic changes by more than 30 seconds. Because it is unclear whether seizures arose from the occipital lesion and spread to temporal regions or whether temporal regions served as primary epileptic foci, the patient undergoes intracranial monitoring. The serial samples show one of the patient’s typical auras of gastric lifting that does not develop into a full complex partial seizure. The patient is awake and signals an aura by pushing an event button. The montage shows intracranial electrodes in the top channels and scalp electrodes in the bottom (intrahippocampal depth electrodes [LHD, RHD]; lesional depth electrode [RPD]; subdural electrodes left frontal [LFS]; subdural electrodes left temporal [LTS]; left occipital [LOS]; right frontal [RFS]; right temporal [RTS]; right occipital [ROS]). Questions:

What is the ictal discharge? What class seizure is this event?

179

Answers: This ictal discharge consists of bursts of rhythmic spikes (first sample) at contact RHD3 (right hippocampal depth electrode). It evolved to rhythmic alpha frequency spikes (second sample) that then slowed to rhythmic theta frequencies (third sample) before slowing further to delta activities before terminating, leaving postictal slowing and suppression within the right hippocampus. The scalp EEG during discharge shows eye movement artifact (first and second samples) and some nonspecific theta slowing (third sample). The ictal discharge briefly spread to the left hippocampus (LHD, third sample). This is a simple partial seizure, an ictal discharge that does not involve enough brain to impair consciousness. Discussion: An ictal discharge is any paroxysmal, rhythmic EEG activity that interrupts ongoing background activities. Ictal discharges typically evolve in frequency, amplitude, and morphology. Ictal discharges designate epileptic seizures, and, if accompanied by clinical symptoms, are diagnostic for epileptic seizures. A simple partial seizure is an event limited to ‘‘simple’’ or single symptoms that do not affect enough brain to impair consciousness. Auras are warning symptoms that often precede complex partial seizures. Recordings of seizures with the use of simultaneous scalp and intracranial electrodes show that auras are merely simple partial seizures that often spare scalp electrodes. These findings are important because lack of ictal changes recorded on the scalp when consciousness is unimpaired does not rule out the possibility of an epileptic seizure. Voltage of the discharge, the degree of synchrony among neurons involved in its generation, its location and orientation, and the affected surface area all govern the appearance of a cortical potential on the scalp. Earlier modeling experiments estimated that 20–70% of spikes recorded from the cortex never appear on scalp electrodes and that at least 6 cm2 (about 1 square inch) of cortex is needed to generate a potential that will be ‘‘seen’’ by scalp electrodes. More recent comparisons of simultaneously recorded intracranial and scalp electrodes in epilepsy surgery candidates demonstrate that 90% of cortical IEDs that emanate from source areas >10 cm2 reach the scalp. Only 10% of cortical IEDs from smaller cortical areas produced scalp potentials.

Despite their apparent accuracy, use of intracranial electrodes is limited to the localization of epileptic foci and to cortical mapping of brain functions. This is not only because of the expense and potential morbidity of an invasive procedure, but also because of the electrical properties of intracranial electrodes. Intracranial electrodes are placed upon or within cortical generators. The small conductive surface of intracranial electrodes results in high impedances. The combination of large signal and large impedances ensures that small sensitivities are required to prevent ‘‘clipping’’ or distortion of the display signal (note the difference in calibration bars between intracranial and scalp recordings in the previous example). Finally, the intensity of electrical fields drops with the square of the distance from the electrical source. The result of these three factors is that intracranial electrodes record from a restricted field; one pair of electrodes may record totally different activities than an immediately adjacent pair. The implication of the limitations inherent in intracranial electrodes is that a great foreknowledge of where seizures will occur is needed in epilepsy surgery planning. Intracranial electrodes cannot be used for diagnosis of nonepileptic pseudoseizures, because the lack of electrographic seizures can be always attributed to electrode placement rather than to a nonepileptic condition. This patient continued to have both auras and complex partial seizures that consistently arose from the right hippocampus. Right anterior temporal lobectomy significantly improved the intensity and frequency of seizures.

Clinical Pearls 1. Auras are functionally simple partial seizures. 2. Because simple partial seizures spare consciousness, and auras and simple partial seizures may not appear on scalp recordings, a normal EEG during an aura or simple partial seizure does not rule out the possibility of epileptic seizures. 3. Intracranial recordings are used for localization before some cases of epilepsy surgery and for cortical mapping of brain function. 180

Ictal Discharges and Epileptic Seizures

REFERENCES 1. Abraham K, Ajmone-Marsen C: Patterns of cortical discharges and their relation to scalp EEG. Electroencephalogr Clin Neurophysiol 1958; 10:447–461. 2. Cooper R, Winter A, Crow H, Walter W: Comparison of subcortical and scalp activity using chronically indwelling electrodes in man. Electroencephalogr Clin Neurophysiol 1965; 18:217–228. 3. Tao JX, Ray A, Hiawes-Ebersole JS, Ebersole JS: Intracranial EEG substrates of scalp EEG interictal spikes. Epilepsia 2005; 46:667–676.

Ictal Discharges and Epileptic Seizures

181

PATIENT 58 A 39-year-old man with frequent spells of right face twitching and tremulousness

A 39-year-old man has recent-onset spells of diffuse tremulousness and loss of consciousness. He is currently taking phenytoin. The recording was a baseline before diagnostic intensive monitoring with video-EEG. Sequential 4-second samples are shown. The technologist asks him questions when she notes the onset of electrographic activity, and the patient responds normally. No motor symptoms are seen. Question: What is the diagnosis of the activity?

182

Ictal Discharges and Epileptic Seizures

Answer: Subclinical rhythmic electrographic discharge in adults (SREDA). The finding is not a seizure and is not associated with epilepsy. Discussion: SREDA is a benign, rhythmic pattern. Although rare, it is important to recognize because it is a rhythmic activity that can be mistaken for an epileptic seizure. Location. The distribution of rhythmic activity is usually diffuse with temporal predominance. Morphology. SREDA consists of the abrupt appearance of rhythmic, monophasic, sharply contoured activities that progressively occur at shorter intervals, usually in the theta frequency band. The progression from slow to fast frequencies is helpful in its distinction from most epileptic ictal discharges, because the latter may evolve from faster to slower frequencies. Reactivity/state. It occurs in adults during restful wakefulness or drowsiness. SREDA is the

exception to many benign epileptiform transients because it usually occurs during wakefulness. Patients have no symptoms during SREDA. Unlike many ictal discharges, there is no postictal slowing of the tracing following resolution. This patient reported none of his typical symptoms during the discharge during the sample. He did, however, experience several of his typical spells during intensive video-EEG. Events featured preserved alpha rhythm during apparent unresponsiveness, a finding not consistent with encephalopathy or ongoing epileptic seizure. His diagnosis was psychogenic nonepileptic pseudoseizures.

Clinical Pearls 1. SREDA is a benign, subclinical burst of rhythmic activity seen in adults that can be distinguished from ictal discharges by its lack of clinical accompaniment and its evolution atypical from seizures. 2. Direct diagnosis of seizures requires the recording of the patient’s typical symptomatology during EEG. Events atypical for the complaints at hand may not predict etiology of chronic, recurrent spells. REFERENCE 1. Westmoreland B, Klass D: A distinctive, rhythmic EEG discharge of adults. Electroencephalogr Clin Neurophysiol 1981; 51:186–189.

Ictal Discharges and Epileptic Seizures

183

PATIENT 59 A 53-year-old man with intractable epilepsy undergoing intensive video-EEG

A 53-year-old man has a 20-year history of medically intractable complex partial seizures and has suspected mesial temporal lobe epilepsy (MTLE). The patient is awake during this recording and eating. Anticonvulsant medications have been withdrawn. Questions: What is the bitemporal rhythmic activity during the first third of the sample? From what region does the seizure arise? Referring to the times at the top of the tracing, how long after clinical onset does electrographic onset of his typical complex partial seizure begin?

184

Ictal Discharges and Epileptic Seizures

Answers: The bitemporal rhythmic activity during the first third of the sample is chewing artifact. The seizure arises from the right anterior-temporal region. Clinical onset occurs at 12:37:14, corresponding to cessation of ongoing waking activities (cessation of chewing artifact, eating lunch). Electrographic onset, marked by emergence of rhythmic fast activities, occurs around 12:37:11. Electrode T8 appears to be the earliest electrode involved, with rhythmic 7 cps sharp theta activity appearing with maximum negativity at electrodes F8 and T2 immediately after onset. This seizure arises from the right side, with best localization to the right anterior-midtemporal region. Electrographic onset precedes clinical onset by ~3 seconds. Discussion: Ictal discharges from focal epileptic lesions can assume various morphologies when recorded from the scalp, depending on such factors as location, depth, and surface area involved. Two typical patterns are 1. Focal attenuation, low-amplitude rhythmic, sharp, fast activity between 10 and 30 cps that gradually builds in amplitude and decreases in frequency 2. Progression of rhythmic spikes or spikewave complexes that build up in frequency before evolving to higher-amplitude, slowfrequency spikes or spike-wave complexes. Following resolution of the ictal discharge, postictal suppression or slowing can be diffuse or more severe from the originating site. In the presurgical evaluation of subjects with MTLE, several features of the ictal discharge recorded from the scalp correlate well with the gold standard of localization with the use of hippocampal depth electrodes: 1. Electrographic onset before emergence of clear clinical symptoms is reassuring. The

reverse sequence—clinical onset before electrographic onset—implies that seizures may arise from an unseen focus before late seizure activity arises on the scalp. 2. Focal development of anterior temporal rhythmic theta or alpha spike activity (>5 cps) is one of the strongest localizing signs in scalp recordings of MTLE patients. 3. Ictal discharges that remain confined to one region, or at least remain unilateral for at least 5 seconds before spreading contralaterally, are good evidence of focal onset. 4. Focal postictal slowing is a reliable indicator of side of seizure onset, but the finding is present in the minority of scalp recordings of patients with MTLE. Many show diffuse postictal slowing. Concordant findings among scalp IEDs, scalp ictal recordings, neuroimaging, and neuropsychological testing predict excellent results (80–90% seizure-free outcome in some series) from anterior temporal lobectomy or selective amygdalahippocampectomy in patients with MTLE.

Clinical Pearls 1. Important localizing features of ictal scalp recordings in patients with MTLE include (1) electrographic onset of focal ictal changes before clinical onset; (2) development of a unilateral, focal ictal discharge consisting of rhythmic theta or alpha activity; (3) confinement of the ictal discharge to one hemisphere for at least 5 seconds before contralateral spread; and (4) focal postictal slowing. 2. Bitemporal bursts of rhythmic muscle activity arise from chewing artifact. REFERENCES 1. Risinger MW, Engel J, Van Ness PC, et al: Ictal localization of temporal lobe seizures with scalp/sphenoidal recordings. Neurology 1989; 39:1288–1293. 2. Williamson P, French J, Thadani V, et al: Characteristics of medial temporal lobe epilepsy. II: Interictal and ictal scalp electroencephalography, neuropsychological testing, neuroimaging, surgical results, and pathology. Ann Neurol 1993; 34(6):781–787.

Ictal Discharges and Epileptic Seizures

185

PATIENT 60 A 30-year-old woman with spells

A 30-year-old woman is evaluated for epilepsy that did not respond to treatment with phenytoin or carbamazepine. Seizures started 1 year prior after a motor vehicle accident in which she had a minor concussion and two family members were killed. The sample is taken during intensive monitoring with video-EEG. The monitor watcher pushes the event button when the patient ceases speaking to another patient, stares, begins facial grimacing, and is unresponsive to her roommate or the monitor-watcher’s interview over the intercom. Question: Does the tracing provide evidence of epileptic seizure?

186

Ictal Discharges and Epileptic Seizures

Answer: No. The tracing shows bursts of muscle activity that often obscure cerebral activity. Posterior rhythms of wakefulness can be discerned intermittently during the spell. This recording is most consistent with a nonepileptic pseudoseizure. Discussion: This example shows evidence of normal ongoing activities of wakefulness despite clinical unresponsiveness. This recording was corroborated by several other events that featured preserved alpha rhythm during apparent unresponsiveness, a finding not consistent with encephalopathy or ongoing seizure. It is a key observation in the diagnosis of psychogenic nonepileptic pseudoseizures (PNES). Clues to diagnosis of PNES are unrecognized depression, anxiety, dissociative traits, or posttraumatic stress disorder. A history of acute stressors can help in the diagnosis of conversion disorder. Patients with PNES have high incidences of sexual or physical abuse. Spells of PNES frequently last longer than the 1–2 minutes of typical complex partial seizures. Spells of PNES may wax and wane in intensity compared to the monophasic course of complex partial seizures. Despite these differences, elaborateness or peculiarity of behaviors is not predictive of PNES, because complex partial seizures of frontal lobe origin are notorious for bizarre symptomatology. Confirmatory monitoring with the use of intensive video-EEG is usually required for diagnosis. Not only do the various symptoms seen in each disease overlap, but comorbidity of pseudoseizures and epileptic seizures ranges from 25– 75% depending upon study methods. Clinical examination during intensive EEG monitoring is an important part of the diagnosis.

Determination of the degree of impairment during spells may facilitate their proper classification. The table below summarizes possible results from the well-conducted diagnostic intensive monitoring session: Clinical Impairment Yes Yes No

Ictal Discharge No

Epileptic seizure Electrographic seizure

Pseudoseizure Simple partial seizure Pseudoseizure

Circumstances exist in which seizures do not appear on scalp recordings. Muscle or movement artifact may obscure the EEG; clues, such as initial focal slowing or IEDs or postictal slowing, may then be helpful. Partial seizures may not involve a critical amount of brain or may arise from deeper cortical areas and remain occult. Sixteen percent of complex partial seizures confirmed by intracranial electrode recordings may not appear on scalp recordings. In cases in which scalp recordings are unclear or obscured, stereotypy of patient’s behaviors during repeated events may be the only recourse to suggest an epileptic cause. Prolactin levels, a pituitary hormone that undergoes acute elevation after complex partial seizures, may aid in diagnosis in selected cases.

Clinical Pearls 1. Psychogenic nonepileptic seizures require the capturing of typical spells on EEG to make a positive diagnosis because the clinical characteristics and comorbidities between nonepileptic and epileptic seizures are high. 2. Alteration of consciousness with preservation of alpha rhythm and other waking activities is characteristic of nonepileptic pseudoseizures captured on EEG. 3. Events without alteration of consciousness or during which consciousness is not determined may have ambiguous conclusions during EEG monitoring. REFERENCES 1. Devinsky O: Nonepileptic psychogenic seizures: Quagmires of pathophysiology, diagnosis, and treatment. Epilepsia 1998; 39:458–462. 2. Gates JR, Ramani V, Whalen S, Loewenson R: Ictal characteristics of pseudoseizures. Arch Neurol 1985; 42:1183–1187. 3. Pacia SV, Ebersole JS: Intracranial EEG substrates of scalp ictal patterns from temporal lobe foci. Epilepsia 1997; 38(6):642–654.

Ictal Discharges and Epileptic Seizures

187

PATIENT 61 A 2-day-old term infant with right body clonus

A 2-day-old term infant (ECA 40 weeks) is born in respiratory distress. Meconium staining is present. Serum sodium is low at 126 mg/dL. Right body clonus is observed by nursing staff shortly after birth. Medications are ampicillin and gentamicin. The recording is performed with routine neonatal polygraphy at the bedside. No age-appropriate activities are seen; instead, unreactive burst-suppression is present throughout the recording. No clinical seizures occurred during the tracing. Question: What is the activity in the left temporal region?

188

Ictal Discharges and Epileptic Seizures

Answer:

Left temporal focal seizure.

Discussion: The main challenge in evaluating neonatal seizures is the distinction of clinical seizures from nonepileptic, stereotypic, and repetitive movements. A second challenge is the identification of electrographic seizures, a task made more difficult because of properties of the immature brain. Because of incomplete connections within and between cortex and subcortical structures, neonates are unable to sustain generalized electrographic seizures. Instead, clonic, multifocal clonic, and tonic seizures occur that involve groups of muscles in a migratory or asymmetric fashion. Motor automatisms, such as ‘‘jitteriness,’’ bicycling, repetitive sucking or orofacial movements, or apnea, are behaviors that are observed frequently in ill infants and usually do not correspond to ictal discharges. Nevertheless, often the EEG is required to supplement clinical observation. Ictal discharges can take several different patterns based on frequency and morphology, but specific patterns correlate poorly with specific causes or outcome. Ictal patterns can consist of (1) rhythmic runs of spikes, (2) sharply contoured

slow waves or epileptiform complexes, or (3) runs of rhythmic activity of changing frequency and morphology. Each kind can remain in one region, migrate from one region to another, or arise in an independent, multifocal pattern. As a rule, however, neonatal electrographic seizures do not generalize. Two features make recognition of neonatal electrographic seizures difficult. The first is that ictal discharges may involve only one electrode and thereby be attributed mistakenly to artifact. The second is that waveforms may remain monomorphic and occur at a slow rhythm, 0.25–1 cps, in some cases. Because the eye becomes trained to ignore very slow activities reminiscent of movement or galvanic artifacts, potential seizures can be easily missed. One technique to emphasize ‘‘latent’’ evolution of morphology and frequency typical of ictal discharges is to decrease the paper speed to compress slow wave activities. The prognosis of neonatal seizures lies with the underlying cause. Electrographic seizures without clinical accompaniment and invariant, agediscordant background activities indicate poor prognosis of normal intellectual development.

Clinical Pearls 1. Neonatal seizures often consist of monomorphic, focal runs of rhythmic activity that, although often migratory or multifocal, do not generalize. 2. Recognition of neonatal seizures after initial screening with standard paper speed can be augmented by review of suspicious rhythmic activities with slow paper speed. REFERENCES 1. McBride MC, Laroia N, Guillet R: Electrographic seizures in neonates correlate with poor neurodevelopmental outcome. Neurology 2000; 55(4):506–513. 2. Mizrahi EM, Kellaway P: Characterization and classification of neonatal seizures. Neurology 1987; 37:1837–1844.

Ictal Discharges and Epileptic Seizures

189

PATIENT 62 A 1-week-old male infant with apnea

A 1-week-old male infant, born at estimated gestational age 39 weeks, presents with a right parenchymal hemorrhage and apneic spells. The infant is treated with phenobarbital. The 30-second sample is taken from routine neonatal polygraphy at the bedside during quiet sleep. Electrographic seizure activity, in the form of rhythmic spike activity starting in the right frontocentrotemporal region and evolving to right centrotemporal delta activity, starts about 10 seconds before the sample. Questions: What is the finding in the respiratory (nasal thermosister) channel? Is there a significant change in heart rate accompanying the seizure?

190

Ictal Discharges and Epileptic Seizures

Answers: The lack of respiratory air flow defines apnea. No significant change in heart rate is present during this apneic seizure. Discussion: Apnea in the full-term neonate is defined as a cessation of breathing for more than 20 seconds. Obstructive apneas are defined as those in which respiratory effort persists throughout the apnea; respiratory effort ceases in central apneas. Mixed apneas contain apneas with intermittent respiratory effort. Respiratory effort is usually measured by thoracic and abdominal strain gauges, inputs that are absent in this example. The differential diagnosis of apnea in the newborn is broad. Apnea can arise from airway obstruction (obstructive apnea), gastroesophageal reflux, cardiopulmonary disease, or, as in this case, epileptic seizures (central or mixed apnea). Furthermore, idiopathic apnea of central or mixed origin may arise because of immaturity of brainstem and peripheral regulatory systems. Epileptic apneas, in comparison to other causes of apnea, are relatively rare in the full-term infant

and not apparent in the premature infant. Most apneas in the full-term neonate are accompanied by bradycardia, the result of over-vigorous cardioinhibitory reflexes of the immature nervous system. Epileptic seizures, conversely, typically override any compensatory reflex activity in the newborn, so that epileptic apneas are rarely associated with bradycardia. For example, in one study of 112 apneas in 15 neonates (6 of whom were premature), epileptic apneas occurred in 4 of the full-term infants. No apneas were accompanied by bradycardia. In this patient’s case, apneas presented as the sole manifestation of epilepsy, a finding not unusual in previous studies. In the full-term neonate, epileptic seizures should be considered as potential causes of apneic events, even in the absence of other behaviors suspicious for seizures.

Clinical Pearls 1. Neonatal nonepileptic apneas typically occur with bradycardia; epileptic apnea usually occurs without bradycardia. 2. Although epileptic apnea should be considered in the full-term infant, epileptic apneas are basically unreported in the premature infant. REFERENCES 1. Fenichel GM, Olson BJ, Fitzpatrick JE: Heart rate changes in convulsive and nonconvulsive neonatal apnea. Ann Neurol 1980; 7:577–582. 2. Tramonte JJ, Goodkin HP: Temporal lobe hemorrhage in the full-term neonate presenting as apneic seizures. J Perinatol 2004; 24:726–729.

Ictal Discharges and Epileptic Seizures

191

PATIENT 63 A 1-week-old female infant with apnea

A 1-week-old female infant (born at estimated gestational age 39 weeks) presents with recurrent episodes of apnea and oxygen desaturation. Apneic spells were associated with tongue thrusting movements. Medications are famotidine. The 30-second sample is recorded during overnight video-polygraphy with the patient asleep (eyes closed on video). Questions: (bars).

192

In what state is the patient during this sample? Describe the two marked episodes

Ictal Discharges and Epileptic Seizures

Answers: The patient is in active sleep. Brief apneas that are not accompanied by bradycardia or CNS disturbance. These are normal apneas. Discussion: Clinically significant apneas in the neonate either exceed 20 seconds in duration, or, if briefer, are accompanied by oxygen desaturation or bradycardia. The incidence and duration of apnea increase with decreasing gestational age at birth. They normally resolve with increasing maturation; in one study, apneic spells ceased in 92% of infants by 37 weeks ECA and in 98% by 40 weeks. Apneas typically occur during active sleep, as the brief events seen in this case. The causes of apnea and their severity are typically divided into those affecting premature infants versus full-term infants. This idiopathic syndrome, termed apneas of prematurity, results from dysregulation of the cardiopulmonary system because of immaturity of the central nervous system and peripheral baroceptor and chemoceptor responses. Diagnosis is made by identifying possible causes of secondary apnea, mainly sepsis, metabolic abnormalities, and

CNS hemorrhage. Monitoring is typically limited to oxygen saturation and EKG monitors. Although clinically significant apneas typically resolve by term, subtle dysfunction, marked by brief apneas and less-rhythmic breathing, persists in infants, particularly those with small birth weights. In term infants, epileptic seizures are an infrequent but important cause of clinically significant apnea. Epileptic apneas usually arise in conjunction with other seizure behaviors but rarely in isolation from other clinical seizure activities. In this patient’s case, overnight polygraphy captured numerous episodes of brief apneas. None were accompanied by electrographic seizure activity, and none exceeded 20 seconds in duration or were associated with desaturation. Because waking and sleep activities were appropriate for ECA and neither seizures nor IEDs were found, epileptic seizures were removed from consideration in the differential diagnosis.

Clinical Pearls 1. Brief apneas without clinical accompaniments are common in term infants. 2. Clinically significant apneas are common in premature infants, with severity and incidence inversely related to ECA. 3. Apneas are most common during active sleep. REFERENCES 1. Curzi-Dascalova L, Peirano P, Christova E: Respiratory characteristics during sleep in healthy small-for-gestational age newborns. Pediatrics 1996; 97:554–559. 2. Henderson-Smart DJ: The effect of gestational age on the incidence and duration of recurrent apnoea in newborn babies. Aust Paediatr J 1981; 17:273–276.

Ictal Discharges and Epileptic Seizures

193

PATIENT 64 A 37-year-old man with apparent complex partial seizures

An EEG is requested in a 37-year-old man with static encephalopathy and apparent recurrent complex partial seizures that has increased in occurrence since a switch to carbamazepine from an unknown anticonvulsant medication. The EEG is recorded with the patient awake. The notations refer to response testing shown in the bottom channel. Deviations in potential are caused by the technologist pushing a button that creates an audible signal to which the patient is instructed to respond with a button of his own. Question: Is this an interictal or ictal discharge? What type of seizure is the patient experiencing?

194

Ictal Discharges and Epileptic Seizures

Answer: The lack of patient responses during spike-wave discharges and an intact response with resolution of the discharge confirm that this is an ictal discharge, that is, both a clinical and electrographic seizure. The sample shows rhythmic, generalized multispike-wave and spike wave discharges that occur at a rate of 2.5–3.5 cps. This discharge is more consistent with an atypical absence seizure than a complex partial seizure. Discussion: Atypical absence seizures are defined as episodes of inattention, behavioral arrest, and limited motor automatisms or changes in tone. Atypical absences usually last longer than absence seizures but often occur less frequently, making them sometimes difficult to distinguish by clinical observation from complex partial seizures. The hallmark of an atypical absence seizure is generalized spike-wave discharges. Unlike the classic 3 cps spike-wave bursts of absence seizures, frequency of occurrence is usually between 1.5 and 2.5 cps. Multiple spike-wave discharges, or a mix of spike-wave and multiple spike-wave, occur frequently. Background EEGs are usually helpful in that they are usually abnormal in those with atypical absences and usually normal in those with typical absences. The division between interictal epileptiform discharges and clinically apparent ictal discharges is not always clear, especially in the generalized epilepsies. For example, although a myoclonic seizure may be accompanied by a

generalized ictal discharge, the duration of the seizure may be too brief to accurately determine any degree of impairment. The division, however, goes beyond academic interest. Questions regarding the ability to drive, assessments of poor learning in school, and efficacy of treatment regimens often pivot on the determination if a specific discharge is accompanied by clinical impairment of consciousness or cognition. An aid in this decision is response testing. As shown in the example, both the technologist and patient have similar buttons. When a discharge occurs on the EEG, the technologist can rapidly ascertain if call-and-response is impaired. In this patient’s case, the distinction between atypical absence seizures and complex partial seizures is important because these findings can direct anticonvulsant selection. The patient’s medication is changed to a broad-spectrum anticonvulsant because carbamazepine, although suited for partial seizures, may exacerbate generalized seizures.

Clinical Pearls 1. Atypical absence seizures are marked by generalized spike-wave or multiple spike-wave bursts with frequencies less than 3 cps. 2. Response testing is a procedure that can distinguish between interictal and ictal activity by documenting impairment in attention and responsiveness. REFERENCES 1. Perucca E, Gram L, Avanzini G, Dulac O: Antiepileptic drugs as a cause of worsening seizures. Epilepsia 1998; 39:5–17.

Ictal Discharges and Epileptic Seizures

195

PATIENT 65 A 25-year-old man with Lennox-Gastaut syndrome and drop attacks

A 25-year-old man has Lennox-Gastaut syndrome that, until recently, was expressed with frequent generalized tonic-clonic seizures and atypical absence seizures. Over the previous 3 months, he began abruptly falling without warning. No overt seizure activity is seen before or after events. The recording is performed with the patient awake. Medications are valproate and zonisamide. While sitting upright, the patient is observed to abruptly stiffen and slump to the floor. Question: What type of seizure accounts for the patient’s drop attacks?

196

Ictal Discharges and Epileptic Seizures

Answer: The recording is diagnostic for a tonic seizure. Generalized fast activity, appearing as diffuse attenuation, corresponds to the patient’s symptoms of brief hypertonia. Discussion: Epileptic drop attacks can result from generalized or rapid secondary generalization of partial seizures. Tonic seizures, shown here, consist of brief, generalized hypertonic episodes, sometimes called axial spasms. The EEG shows bursts of generalized fast activities, usually preceded by a high-amplitude, generalized spike-wave. Tonic seizures are most commonly encountered in Lennox-Gastaut syndrome; in fact, nocturnal bursts of generalized fast activities accompanied by tonic seizures are thought by some to be pathomnemonic of Lennox-Gastaut syndrome. Atonic seizures cause falls from abrupt generalized loss of tone. Various ictal discharges can be seen. Atonic seizures preceded by brief myoclonus are seen in the rare epileptic syndrome

of childhood myoclonic-astatic epilepsy (Doose syndrome), differentiated from Lennox-Gastaut syndrome by its association with relatively preserved cognition, predisposition to photoparoxysmal responses, and family history. Partial seizures can rapidly secondarily generalize and interrupt maintenance of extensor tone. So-called ‘‘temporal lobe syncope’’ may occur in patients with medically intractable complex partial seizures of either temporal or frontal lobe origin. Syncope from various common cardiovascular causes, as well as rare cases of epileptic seizures that induce bradycardia or asystole, can mimic primary epileptic drop attacks. Cataplexy, in the setting of unrecognized narcolepsy, can be confusing as the cause of drop attacks until the characteristic daytime sleepiness is characterized.

Clinical Pearls 1. Tonic seizures are pathomnemonic findings in Lennox-Gastaut syndrome and consist of bursts of generalized fast activity. 2. The differential diagnosis of epileptic drop attacks includes generalized seizures and rapidly secondarily generalized seizures from frontal or temporal lobe foci. Syncope and cataplexy may present with drop attacks. REFERENCES 1. Brenner RP, Atkinson R: Generalized paroxysmal fast activity: Electroencephalographic and clinical features. Ann Neurol 1982; 11:386–390. 2. Doose H: Myoclonic-astatic epilepsy. Epilepsy Res Suppl 1992; 6:163–168. 3. Gambardella A, Reutens D, Andermann F: Late-onset drop attacks in temporal lobe epilepsy: A reevaluation of the concept of temporal lobe syncope. Neurology 1994; 44(6):1074–1078. 4. Quigg M, Bleck T: Syncope. In Engel J, Pedley T (eds). Epilepsy: A Comprehensive Textbook. New York, LippincottRaven, 1997, pp 2649–2659.

Ictal Discharges and Epileptic Seizures

197

PATIENT 66 A 17-month-old girl with drop attacks

A 17-month-old girl has onset of recurrent drop attacks 1 month prior. Drop attacks were usually followed by brief episodes of myoclonic jerks. The child was born at term and has normal development. Although the parents state that spells occurred spontaneously, on closer history the child was often crying before the episodes. The spells did not resolve with a trial of phenytoin. The recording is performed with the child awake, upset, and crying. She is taking no medications. Question: What is the etiology of the event captured during the tracing?

198

Ictal Discharges and Epileptic Seizures

Answer: The tracing shows abrupt development of generalized, semirhythmic delta activity followed by suppression coinciding with onset of symptoms. The EKG shows loss of the QRS complex before EEG changes. This recording is diagnostic for cerebral hypoperfusion caused by crying-provoked asystole (breath-holding). Discussion: Syncope refers to the abrupt and transient loss of consciousness and motor tone caused by loss of cerebral perfusion. The most common causes stem from alterations of cardiovascular tone or direct cardiac dysfunction. Breath-holding events usually occur in infants and toddlers between ages 6 and 18 months. Spells are usually triggered by crying but can also be provoked by grunting or defecation or other activities that raise intrathoracic pressure. Traditionally, spells are divided into cyanotic and pallid subtypes. Transient cerebral hypoperfusion causes abrupt changes in the EEG. Usually, bursts of semirhythmic-to-rhythmic delta activities emerge with loss of consciousness and can be followed by suppression if lasting for more than 5–10 seconds. Syncopal myoclonus, sometimes termed syncopal

convulsions, consisting of brief, generalized myoclonic jerks, is present in over 80% of cases of syncope induced in research studies. No ictal discharges accompany syncopal myoclonus. EEG is not a useful screening tool in most cases of routine syncope. Nevertheless, EEG is called upon during evaluations of more unusual or severe cases. In this case, syncope was suspected during initial consultation because drop attacks from epileptic causes usually occur in syndromes with abnormal developmental histories, for example, tonic seizures in LennoxGastaut syndrome. This child was managed conservatively at first, but spells recurred. Cardiology evaluations were normal. Finally, two severe episodes lasting over 2 minutes occurred, and her cardiologist recommended placement of a cardiac pacemaker.

Clinical Pearls 1. The EEG during syncope shows development of generalized delta activity corresponding to acute loss of consciousness. Progression to suppression may occur if cerebral hypoperfusion persists. 2. Syncopal myoclonus can be observed during many episodes of syncope and are not epileptic in origin. 3. Although EEG is not a useful screening tool in routine evaluation of syncope, in unusual cases it is a sensitive and specific technique, if a spell is captured. REFERENCES 1. Aminoff M, Scheinman M, Griffin J, Herre J: Electrocerebral accompaniments of syncope associated with malignant ventricular arrhythmias. Ann Intern Med 1988; 108(6):791–796. 2. Quigg M, Bleck T: Syncope. In Engel J, Pedley T (eds). Epilepsy: A Comprehensive Textbook. New York, LippincottRaven, 1997, pp 2649–2659.

Ictal Discharges and Epileptic Seizures

199

EFFECTS OF FOCAL LESIONS ON EEG 11

PATIENT 67 A 73-year-old woman with confusion and left hemiparesis

A 73-year-old woman is admitted with a right hemispheric intracerebral hemorrhagic stroke causing left hemiparesis. After admission, her level of consciousness declines, and repeat neuroimaging shows no evidence of progressive bleed or edema. The patient during this recording is confused but awake. She is taking no CNS-active medications. The patient is asked to close her eyes but only flutters them in response. Questions:

What is the abnormality? What is the most likely cause?

201

Answers: Continuous, unreactive, arrhythmic delta activity is present across right frontaltemporal-central regions. Focal arrhythmic delta activity indicates a structural lesion of the area, corresponding to the region of hemorrhagic stroke. Discussion: At one time, EEG, neurologic examination, arteriography, and pneumoencephalography were the only means by which to determine the existence of a focal cerebral lesion in the living patient. Autopsy was often the tie-breaker in unclear cases. Currently, the neuroimaging techniques of computed tomography (CT) and magnetic resonance imaging (MRI) have largely supplanted EEG’s former role in the investigation of focal lesions. MRI, for example, far exceeds the ability of EEGs to determine the presence of a focal tumor. Nevertheless, the usefulness of EEG exceeds its relatively poor sensitivity, for it currently remains the only widespread tool, outside of clinical exam, that can determine the functional consequences of a presumed lesion. Focal lesions alter the EEG in three important ways. 1. Focal lesions may cause changes in normal activities. For example, the alpha rhythm ipsilateral to the hemisphere containing the focal lesion may demonstrate slowing in frequency, decrement in amplitude, or a defect in persistence and reactivity. Sleep activities may be poorly developed or absent on the affected side. 2. Focal lesions may cause abnormalities elicited with activation procedures. Asymmetry of photic driving suggests the presence of an ipsilateral physiologic or anatomic lesion. A similar interpretation follows focal slowing induced during hyperventilation. These

physiologic changes, however, should be corroborated by other evidence of localized dysfunction. 3. Focal lesions may also cause the emergence of abnormal activities, specifically, focal theta or delta activity. Two main morphologies of slowing are intermittent rhythmic delta activity (IRDA) and arrhythmic delta activity (ADA). IRDA appears as bursts of sinusoidal, rhythmic delta activities, usually reactive to state or stimulation, and implies physiologic rather than structural abnormalities. ADA consists of mixed low frequencies and is usually the consequence of a fixed lesion. The mixture of slow frequencies creates an appearance of ‘‘polymorphic’’ slowing. The technologist should demonstrate whether slowing is reactive and altered or improved with patient stimulation or endogenous arousal. Although there are no clear divisions in the severity of focal slowing, more slow frequencies, greater persistence, and unreactivity correspond to more profound lesions. Studies show that focal slowing is most reliably generated from white matter lesions that interrupt connections to and from the cortex. Tumors, strokes, and abscesses are typical causes of structural lesions. Focal slowing is a nonspecific response; therefore, physiologic lesions, such as postictal slowing from epileptogenic zones or even transient perfusion abnormalities from complicated migraine, can generate focal slowing.

Clinical Pearls 1. Focal slowing indicates localized cerebral dysfunction and most reliably occurs with disruption of the underlying white matter. 2. Focal slowing can take the form of sporadic, intermittent, or continuous slowing. 3. Intermittent rhythmic delta activity implies a physiologic lesion, and arrhythmic delta activity structural lesion, although activities are certainly not restricted to these specific interpretations. REFERENCES 1. Gloor P, Ball G, Schaul N: Brain lesions that produce delta waves in the EEG. Neurology 1977; 27:326–333. 2. Schaul N: Pathogenesis and significance of abnormal nonepileptiform rhythms in the EEG. J Clin Neurophysiol 1990; 7:229–248. 3. Walter WG: The location of cerebral tumours by electroencephalography. Lancet 1936; 2:305–308.

202

Effects of Focal Lesions on EEG

PATIENT 68 A 33-year-old man with suspected mesial temporal lobe epilepsy

A 33-year-old man presents with adolescent-onset, medically refractory complex partial seizures with features suggesting the syndrome of mesial temporal lobe epilepsy (MTLE). Medications are levetiracetam and carbamazepine. The recording is obtained with the patient awake. The sample is shown in both bipolar and referential montages. Bursts of delta activity shown here occur intermittently during the tracing. Questions:

What is the location of focal slowing? What is its clinical import?

Effects of Focal Lesions on EEG

203

Answers: The focal slowing is temporal intermittent rhythmic delta activity (TIRDA), with widespread field with highest amplitudes across the temporal region. It is potentially epileptogenic. Discussion: TIRDA occupies a special place in the hierarchy of focal delta activity. Unlike other focal slowing, a nonspecific indicator of localized cerebral pathology, TIRDA is associated with epileptogenic regions in patients with epilepsy. Thus, TIRDA is the only epileptogenic focal finding that is not epileptiform. TIRDA consists of intermittent bursts of sinusoidal delta activities. Sharp waves or spikes are often present in the same tracing and may appear with bursts of TIRDA. TIRDA has been usually studied in the setting of MTLE during consideration of epilepsy surgery. In this subgroup of patients, TIRDA is highly

predictive of side of seizure onset and thus a finding supporting lateralization and localization. TIRDA is not limited to patients with MTLE and has been seen in patients with localization-related epilepsies not of temporal lobe origin. Nevertheless, its close association with localization-related epilepsies makes TIRDA an epileptogenic finding. Sporadic temporal slowing, in contrast, does not have the same specificity of TIRDA in MTLE and is not epileptogenic. To maintain specificity of TIRDA compared to sporadic slowing, many conservatively reserve TIRDA for activity that truly appears only in rhythmic bursts and is recurrent during the recording.

Clinical Pearls 1. TIRDA is focal, intermittent, rhythmic delta activity that is potentially epileptogenic. It is the best documented epileptogenic finding that is not epileptiform. 2. TIRDA in suspected MTLE is highly predictive of side of seizure onset. REFERENCES 1. Geyer JD, Bilir E, Faught E, et al: Significance of interictal temporal lobe delta activity for localization of the primary epileptogenic region. Neurology 1999; 52:202–205. 2. Normand MM, Wszolek ZK, Klass DW: Temporal intermittent rhythmic delta activity in electroencephalograms. J Clin Neurophysiol 1995; 12:280–284.

204

Effects of Focal Lesions on EEG

PATIENT 69 A 30-year-old man with complex partial seizures

An EEG is requested in a 30-year-old man who experienced an exacerbation of complex partial seizures despite documented anticonvulsant medication compliance. Medications are carbamazepine and phenytoin. The patient is awake during the recording. Question:

What are the location and source of focal delta activity?

Effects of Focal Lesions on EEG

205

Answer: Frequent slow-wave discharges are present in the left frontotemporal channels. The source is left eye movement artifact in a subject with an enucleated right eye. Discussion: Admittedly, the EEG technologist should document any usual physical features of the patient, such as skull defects and scars, or, as in this case, an enucleated eye. Even in the absence of eye leads, eye movement artifact can usually be distinguished from

frontal delta activity. Eye movement artifact is typically confined to the first two channels of a longitudinal montage, whereas most frontal delta activity of brain origin is broadly distributed.

Clinical Pearls 1. Physical characteristics of the patient must be documented by the technologist. 2. Eye movement artifact is usually confined to frontopolar and frontotemporal fields.

206

Effects of Focal Lesions on EEG

PATIENT 70 A 28-year-old woman with psychic auras

A 28-year-old woman presents with feelings of dissociation and de´ja vu that have been occurring for the previous 2 years. She is taking no medications. The recording is made with the patient awake. The sample is shown in bipolar longitudinal and right-ear referential montages. Question:

What is the finding in the diagram?

Effects of Focal Lesions on EEG

207

Answer:

Left anterior midtemporal theta activity.

Discussion: Sporadic, subtle slowing is a recurrent problem in interpretation. In adults, slowing in the temporal regions is particularly vexing, and clinical significance varies with the type of slowing encountered and the age of the patient. The most common of ambiguous slowings are brief bursts of low to medium amplitude (10 cm (40% interelectrode distances in the 10–20 system). An increased interelectrode distance amplifies possible brain signal. Calibration. To confirm the integrity of signal from patient to pen, the technologist taps each electrode in turn to record the resulting artifact. Electrode impedances must be within standard limits.

Sensitivity and duration. Recording at 2 mV/ mm (or its digital peak-to-peak equivalent) for at least 30 minutes is required. The high sensitivity represents the threshold below which cerebral activity at the scalp is indistinguishable from noise. Reactivity. Response testing to painful stimuli is mandatory. Artifact and filters. Artifact must be identified and eliminated by the technologist, and whatever artifact remains must be identified. The ICU is rich with electrical noise, but it can usually be eliminated to a satisfactory degree. More problematic is persistent EKG artifact. Although the QRS wave is easily identifiable, the T wave can appear as rhythmic slow-wave activity. Rhythmic pulsatile artifact from underlying scalp blood flow and IV pumps may be present. Vexing, periodic artifacts, such as filling of airflow beds, deep venous thrombosis stockings, and ventilator vibration, can all appear as possible EEG bursts. Filters cannot be adjusted beyond 1 and 30 Hz. Reversible causes of ECS include severe drug overdose and hypothermia; therefore, EEG cannot augment clinical exam in those conditions. Reversible ECS may occur during shock or other causes of cerebral hypoperfusion. Aside from these potential confounders, electrocerebral silence confirms a clinical diagnosis of brain death.

Clinical Pearls 1. ECS, when recording with accepted cerebral death examination protocols, denotes absence of cerebral activity and supports a diagnosis of brain death. 2. CDE protocol includes increased interelectrode distances, sensitivities of 2 mV/ mm for at least 30 minutes, and identification and elimination of artifact. REFERENCE 1. American Clinical Neurophysiology Society: Guideline 3: Minimum technical standards for EEG recording in suspected cerebral death. J Clin Neurophysiol 2006; 23:97–104.

248

Effects of Diffuse Encephalopathy on EEG

PERIODIC EPILEPTIFORM DISCHARGES 13

PATIENT 90 A 61-year-old woman with metastatic melanoma and obtundation

A 61-year-old woman with metastatic melanoma presents with acute worsening of level of consciousness that has progressed over the previous 48 hours. On examination she is minimally responsive to tactile stimulation. A head CT with contrast shows meningeal enhancement of the tentorium and lack of hydrocephalus. She is on no medications. The EEG is performed to rule out possible seizure activity. Question:

What are the repetitive discharges apparent in the posterior head region?

249

Answer: Quasi-periodic, triphasic, and polyphasic complex discharges are present, some of which appear epileptiform, with a period of recurrence at about every 3–4 seconds. They are bioccipital periodic epileptiform discharges (PEDs). Discussion: Periodic discharges consist of waveforms that stand out from and interrupt background activity in a recurrent, regular pattern. Cortical activities are hardly ever truly periodic; artifact from biologic sources, such as EKG and from exogenous sources as ventilators, are more likely sources. Instead, quasi-periodic denotes the more typical pattern of cortical origin with a range of timings that separate discharges. For brevity, most refer to periodic discharges despite the important distinction. PEDs are sharp transients that recur in a periodic fashion. Frequency and timing. The distinction between PEDS and the bursts of suppression-burst patterns is important. Bursts in suppression-bursts, like PEDs, can certainly occur periodically. Bursts, however, occur on the relative absence of background activities; PEDs usually interrupt background activities. PEDs, although often complex in morphology, imply discharges that last at the longest 1–1.5 seconds, whereas the briefest bursts in suppression-burst usually exceed 1–2 seconds in duration. Although the timing between PEDs is variable, within an individual study the range of timings is fairly constrained. Typical PEDs recur every 1–2 seconds, with extremes between 0.5 and 5 seconds. Although epileptiform discharges may recur periodically, a brief train of periodic discharges is not sufficient for the designation PEDs. The term should be reserved for situations in which the discharges are continuous and invariably present throughout a tracing; a 10-minute minimum has been applied in some studies. Location. PEDS can be generalized, dominant in one region (periodic lateralized epileptiform discharges [PLEDs]), or occur independently or dependently in homotopic distributions (bilateral periodic discharges [BiPEDs]). These distinctions aid in description, but localized pathologies are not necessarily constrained to the production of PLEDS. Localized and multifocal lesions may both produce PEDs or PLEDs. Morphology. The morphology of PEDs varies widely, ranging from sharply contoured slow wave discharges to spikes to complex, polyphasic sharp bursts. Despite the wide range of morphol-

250

ogies among studies, PEDs within the same study are similar (but not identical) and repetitive. The emergence, morphology, and interdischarge interval of PEDs varies with the course and duration of the underlying cause. PEDs can emerge transiently. They appear most reliably within a day or so of the acute injury and are at their highest amplitude and complexity early. The duration of each discharge tends to be at its shortest early on. With time, the interdischarge interval increases and complexity decreases. Eventually, PEDs are usually replaced by arrhythmic delta activity, the familiar sign of chronic structural injury. Reactivity. PEDs tend to resist attenuation with external stimuli or with endogenous state. PEDs usually occur in addition to other EEG signs of encephalopathy, so reactivity of background activities varies with the depth of loss of consciousness. PEDs usually indicate an acute or subacute structural lesion of grey or white matter, typically both. Acute stroke is the most common cause of PEDs. CNS infections are another common cause, with herpes simplex encephalitis and Creutzfeldt-Jakob disease being other traditionally mentioned specific causes. Occasionally, metabolic disorders will provoke PEDs in those with existing localized pathology. Seizures commonly coexist with PEDs. Even though seizures and PEDs seemingly go hand-inhand, PEDs are not necessarily predictive of future epileptic seizures. Seizures and PEDs probably represent the coexisting signs of acute brain injury, rather than evidence of the future risk of seizures. In other words, PEDs are an epileptiform pattern that is not clearly epileptogenic. An important refinement to that last statement is that PEDs may indicate ongoing seizure activity. PEDs occur in end-stage status epilepticus, in experimental animal models of epilepsy, and may represent the ictal discharge in patients with nonconvulsive status epilepticus. In this patient’s case, the EEG finding of bioccipital PEDs leads physicians to acquire an MRI that demonstrates bilateral posterior invasion of metastatic lesions. Malignant and rapid invasion of tumor is another etiology associated with PEDs.

Periodic Epileptiform Discharges

Clinical Pearls 1. PEDs are continuously present, poorly reactive, periodic epileptiform discharges. 2. PEDs usually denote acute destructive lesions. 3. Although seizures may occur during the acute brain lesion that causes PEDs, PEDs are not predictive of future risk of epileptic seizures. 4. PEDs are one of the forms of chronic or ongoing ictal discharges and can be seen in nonconvulsive status epilepticus. REFERENCES 1. Chong DJ, Hirschl J: Which EEG patterns warrant treatment in the clinically ill? Reviewing the evidence for treatment of periodic epileptiform discharges and related patterns. J Clin Neurophysiol 2005; 22:79–91. 2. Garcia-Morales I, Garcia MT, Galan-Davila L, et al: Periodic lateralized epileptiform discharges: Etiology, clinical aspects, seizures, and evolution in 130 patients. J Clin Neurophysiol 2002; 19(2):172–177. 3. Pohlmann-Eden B, Hoch DB, Cochius JI, Chiappa KH: Periodic lateralized epileptiform discharges—A critical review. J Clin Neurophysiol 1996; 13(6):519–530.

Periodic Epileptiform Discharges

251

PATIENT 91 A 70-year-old man with fever, confusion, and aphasia

An EEG is requested for a 70-year-old man with several days of fever followed by confusion and expressive aphasia. Medications include broad-spectrum antibiotics and acyclovir. The recording is made with the patient awake. Questions: What are the EEG findings? What is the most likely diagnosis given the clinical information and EEG findings?

252

Periodic Epileptiform Discharges

Answers: Background activities consist of arrhythmic delta and theta activities. PLEDs are present with greater involvement of the left hemisphere with phase-reversal variably present in the posterior temporal region. The EEG supports a diagnosis of acute herpes simplex viral encephalitis. Discussion: Historically, EEG was an important tool in the diagnosis of possible herpes simplex viral encephalitis (HSVE). PLEDs, in the setting of fever and delirium or lethargy, occurred earlier and was a more specific finding than changes on head CT in early HSVE. PLEDs, therefore, were evidence that often paved the way to a diagnostic brain biopsy. The availability of effective therapy for HSVE, however, modified the need for acute EEG in the diagnostic and treatment plan. Currently, patients with suspected HSVE are routinely treated with the antiviral agent acyclovir while awaiting detection of HSV in the plasma or cerebrospinal fluid (CSF) through the use of the polymerase chain reaction (PCR) test. Nevertheless, EEG does have an adjunctive and useful role in the current process. First, the findings of PLEDs in suspected HSVE remain a sensitive test. For example, in

neonates with HSVE, PLEDs or other focal abnormalities on EEG are present in over 80%, compared to other signs, such as HSV rash (40%) or MRI abnormalities (65%) within 12 days of clinical onset. PLEDs may have a prognostic value in PCRproven HSVE. PLEDs are associated with worse neurologic outcome. Past studies also showed that abnormal background activities in conjunction with PLEDs also correlated with poor neurologic outcome, but this later finding may not hold true in the age of acyclovir treatment. In this patient’s case, the PCR test was positive for HSV. The left temporal PLEDs were found in the third day after clinical onset of fever. A subsequent EEG repeated on hospital day 10 (13 days after onset) showed left temporal arrhythmic delta activity. Repeat EEGs are recommended to document the gradual evolution of PLEDs.

Clinical Pearls 1. PLEDs are an important adjunct in the evaluation and treatment of herpes simplex viral encephalitis. 2. PLEDs appear earlier in the time course of herpetic encephalitis than neuroimaging. 3. The persistence and presence of PLEDs in herpetic encephalitis are associated with worse outcome in acyclovir-treated patients. REFERENCES 1. Kimberlin DW, Jacobs RF, Powell DA, et al: Natural history of neonatal herpes simplex virus infections in the acyclovir era. Pediatrics 2001; 108:223–229. 2. Siren J, Seppalainen AM, Launes J: Is EEG useful in assessing patients with acute encephalitis treated with acyclovir? Electroencephalogr Clin Neurophysiol 1998; 107(4):296–301.

Periodic Epileptiform Discharges

253

PATIENT 92 A 71-year-old man with rapidly progressive memory loss and somnolence

A 71-year-old man presents with rapidly progressive memory loss, personality changes, and somnolence. He has a history of bipolar disorder and chronic obstructive pulmonary disease (COPD). He is noted to have intermittent shaking of the left arm. Medications include lithium and valproate. The EEG is performed with the patient awake. There are no EEG correlates to left arm movements. Questions: What is the predominant EEG finding? What group of encephalopathies does this sample suggest in the setting of rapidly progressive dementia?

254

Periodic Epileptiform Discharges

Answers: PLEDs are present in the left centrotemporal region, occasionally involving the contralateral hemisphere. In the setting of rapidly progressive dementia, PLEDs suggest prion infection (Creutzfeldt-Jakob disease). Discussion: The most common human prion disease is Creutzfeldt-Jakob disease (CJD), which accounts for about 85% of prion infections. The primary symptoms of CJD are a rapidly progressive dementia, myoclonus, and variable evidence of multifocal neurologic disease. The EEG in CJD typically shows PEDs or PLEDs on abnormally slow background activity. Myoclonus coincides with PEDs, but treatment with anticonvulsants does not necessarily resolve either. The EEG has reasonable specificity and sensitivity (67% and 87%, respectively) in predicting the typical spongiform degeneration seen on autopsy or biopsy. The main shortcoming of EEG in the diagnosis of CJD is that, like the usual course of PEDs in structural injuries, PEDs occur transiently and may be absent in early stages of the disease. PEDs occur about 2 months after the first clinical changes, usually coinciding with the development of myoclonus and akinetic mutism. One study found that FIRDA often preceded onset of PEDs and myoclonus and should guide recommendations for repeat study. Repeat recordings

can greatly improve the chances of a positive diagnosis, so that 90% of patients with CJD at one point of their illness demonstrate PEDs. Other prion diseases lack a clear association with PEDs, so that lack of PEDs does not provide evidence against such variants as fatal familial insomnia or the human form of bovine spongiform encephalopathy. One concern is the iatrogenic spread of prion disease with the reuse of EEG electrodes. Because the prion protein is resistant to common, vigorous methods of sterilization, EEG electrodes are not reused if the clinical question is possible CJD. An exception to this rule is that electrodes may be autoclaved and bleached for later reuse for an individual patient. Other infectious agents besides CJD can cause PEDs. Subacute sclerosing panencephalitis (SSPE) is a chronic measles encephalitis of childhood that is now rare in the United States because of immunization. As in CJD, PEDs appear in the setting of dementia and myoclonus. Sometimes the interval between periodic complexes can be quite prolonged in SSPE.

Clinical Pearls 1. Findings of PEDs in the setting of dementia and myoclonus are suggestive of prion disease. 2. Electrodes are not reused if the differential diagnosis is prion disease. 3. In children, SSPE should join the differential diagnosis, especially if the intercomplex interval is prolonged. REFERENCES 1. Aguglia U, Farnarier G, Tinuper P, et al: Subacute spongiform encephalopathy with periodic paroxysmal activities: Clinical evolution and serial EEG findings in 20 cases. Clin Electroencephalogr 1987; 18(3):147–158. 2. Chiofalo N, Fuentes A, Galvez S: Serial EEG findings in 27 cases of Creutzfeldt-Jakob disease. Arch Neurol 1980; 37(3):143–145. 3. Hansen HC, Zschocke S, Sturenburg HJ, Kunze K: Clinical changes and EEG patterns preceding the onset of periodic sharp wave complexes in Creutzfeldt-Jakob disease. Acta Neurol Scand 1998; 97(2):99–106.

Periodic Epileptiform Discharges

255

PATIENT 93 An 83-year-old woman with recurrent confusion and seizures

An emergent EEG is requested to evaluate delirium in an 83-year-old, right-handed woman with a history of an old left cerebral hemorrhage. She presents with a seizure involving the right arm and the face. She was seen a week before in the emergency department with confusion that spontaneously improved with no evidence of metabolic abnormalities or new stroke. Symptoms recurred intermittently since. On this admission, she cannot follow commands on exam. The EEG is obtained in the emergency department with the patient awake. Medications are unknown. Questions: What are the two locations of periodic discharges in this sample? Does this recording refute the diagnosis of delirium?

256

Periodic Epileptiform Discharges

Answers: PLEDs with a frequency of 1.5–2 cps occur broadly across the left anterior temporal region. Periodic activity across the right temporal region arises from EKG artifact. The relatively fast activities apparent in background activity argue against a metabolic encephalopathy. ‘‘Delirium’’ in this case may be referable to a receptive aphasia rather than a bihemispheric encephalopathy. Discussion: This patient’s case illustrates the use of EEG in aiding the diagnosis of neurologic syndromes that can present with similar presentations. In this case, waxing and waning attention and inability to follow commands could result from either a diffuse encephalopathy causing delirium or focal dysfunction of the dominant hemisphere inducing a receptive aphasia. This tracing shows prominent left temporal PLEDs and independent right temporal periodic discharges. The EKG channel, however, demonstrates that the latter are EKG artifact rather than independent PLEDs. Another hint for proper distinction between the two is that left temporal PLEDs occur in a quasi-periodic pattern every 1.5–2 seconds, but right temporal artifact is truly periodic at the cardiac sinus rhythm. The background activities consist of poorly organized theta activities across the left hemisphere and 9–10 cps, posteriorly dominant alpha activities across the right hemisphere (with occasional low-amplitude, posteriorly dominant alpha

activities apparent on the left as well). Although alpha rhythm is not explicitly demonstrated with eye opening or closure, in this sample, the frequencies and distribution can be considered presumptive alpha rhythm. Left temporal PLEDs and mild slowing of background activities of wakefulness across the left hemisphere suggest acute localized dysfunction. The findings here are not specific enough to determine whether her symptoms are referable to an acute destructive lesion, such as new stroke, an ongoing complex partial seizure (complex partial status epilepticus), or a combination of the two. This patient’s syndrome spontaneously resolved several hours after the EEG. A head CT showed no evidence of new infarct or hemorrhage. With the differential diagnosis, including transient ischemic attacks and recurrent complex partial seizures/nonconvulsive status epilepticus, she was treated with antiplatelet therapies and anticonvulsant medications (carbamazepine).

Clinical Pearls 1. Accurate interpretation of periodic discharges requires the use of an EKG channel. 2. PLEDs can represent an acute structural lesion or ongoing seizure activity. REFERENCE 1. Chong DJ, Hirschl J: Which EEG patterns warrant treatment in the clinically ill? Reviewing the evidence for treatment of periodic epileptiform discharges and related patterns. J Clin Neurophysiol 2005; 22:79–91. 2. Kaplan PW: Nonconvulsive status epilepticus in the emergency room. Epilepsia 1996; 37(7):643–650.

Periodic Epileptiform Discharges

257

STATUS EPILEPTICUS 14

PATIENT 94 A 52-year-old man with epilepsy found inattentive and disoriented

An EEG is requested to evaluate possible ongoing seizure activity in a 52-year-old man with a history of idiopathic generalized seizures who was found confused. The patient is taking phenytoin and valproate. Two 20-second samples are shown. During the first, baseline sample (‘‘before diazepam’’), the patient is awake, inattentive, and disoriented; during the second (‘‘after diazepam’’), the patient is more alert and attentive but remains disoriented. Questions: Describe the baseline EEG. Do the responses to the ongoing treatment confirm a diagnosis of nonconvulsive status epilepticus?

259

Answers: The baseline EEG depicts recurrent bursts of rhythmic multiple spike discharges on background activities of diffusely distributed alpha activity. Bursts of multiple spike activity resolve coinciding with clinical improvement. Resolution of clinical impairment and electrographic seizure activity following anticonvulsant treatment is diagnostic of nonconvulsive status epilepticus (NCSE). Discussion: NCSE is a state of continuous electrographic seizures lasting longer than 30 minutes that is not accompanied by clinically obvious motor activity. Instead, a wide range of impairments of consciousness accompanies electrical status epilepticus, from a vague, subjective discomfort, to impaired attention, lethargy, waxing and waning delirium or clouded consciousness, stupor, or coma. Sometimes simple, repetitive, and subtle motor activities are present, such as beating lateral nystagmus-like activity, facial or periorbital clonus, repetitive posturing, or changes in tone. Henri Gastaut is attributed with the observation that there are as many different types of status epilepticus as there are types of seizures. Generalized convulsive status epilepticus (CSE), although the most common of status epilepticus syndromes, is mainly a clinical diagnosis; EEG is not required, and, in field conditions, is often not emergently available. However, 14–20% of patients with CSE continue to have EEG evidence of ictal activity (i.e., NCSE) despite resolution of clinical signs of seizures. The role of 260

EEG in CSE, therefore, is to confirm the resolution of ongoing seizure activity in those patients who fail to improve following cessation of clinical convulsions and to monitor the success of ongoing therapy. In this case, the patient had known idiopathic generalized epilepsy, emphasizing that NCSE can be divided in two basic categories: Absence status epilepticus (i.e., ictal stupor, spike-wave stupor) consists of prolonged episodes of absence seizures that are accompanied by continuous generalized discharges. It is usually seen in patients with idiopathic generalized epilepsies. Beyond the morbidity of impaired consciousness, outcome is thought to be relatively benign. Partial status epilepticus (i.e., subtle status epilepticus, complex partial status epilepticus) consists of prolonged focal or regional ictal discharges of various morphologies. Secondary generalization may occur. In contrast to absence status epilepticus, acute brain injury may cause seizures, and outcome is often tied to the underlying etiology. In addition, complex partial status Status Epilepticus

epilepticus in experimental animal models causes permanent neuronal injury. In humans, complex partial status epilepticus is associated with subsequent cognitive impairment. Although EEG is required to make the diagnosis of NCSE, the ongoing ictal discharge often does not cleanly fall into localized or generalized seizure patterns, and one may evolve into the other during prolonged episodes. Adding to the challenge is that ictal discharges are often difficult to distinguish from other rhythmic, nonictal activity, such as PEDs or triphasic waves. Patients who meet the following criteria have definite NCSE: 1. Clinical state is impaired from baseline and is present continuously or intermittently without full recovery for 30 minutes or more. 2. EEG shows ongoing electrographic seizure activity. 3. Interictal or postictal EEG shows IEDs. 4. Both impaired clinical state and electrographic seizure activity improve or resolve following treatment with anticonvulsant medications. Of these criteria, the last is often the most difficult to fulfill. First, clinical improvement may be subtle, so that preictal and postictal testing should strive to document consistent observations in attention, level of consciousness, and activities.

Second, postictal state can be prolonged following status epilepticus. Outside of those patients with absence status epilepticus after which recovery can be instantaneous, clinical improvement lags behind electrographic resolution. Third, the underlying etiology of status epilepticus may independently cause impairment from seizure activity, so that the states of comatose patients rarely change in the course of a routine EEG. Serial or continuous prolonged EEGs, supplemented with clinical examination, may, with time, demonstrate clinical improvement. Drug administration in the diagnosis of NCSE is often out of the hands of the interpreting electroencephalographer, but the team of physicians should be aware that medication and administration may aid in a clear diagnosis. The agents most useful for acute administration during EEG in the diagnosis of NCSE are the benzodiazepines. Lorazepam is the favored agent because of efficacy demonstrated in the multicenter study of treatment of CSE. Other electroencephalographers prefer diazepam or midazolam for diagnosis, rather than treatment because of more rapid dispersion into the CNS. Whichever the specific benzodiazepine, it should be administered intravenously and flushed rapidly to induce a quick and unambiguous change in the ongoing EEG. Consistent clinical testing establishes the patient’s best performance before and after empiric treatment.

Clinical Pearls 1. NCSE consists of continuous electrographic seizure activity with the clinical accompaniment of impairment or loss of consciousness. 2. Partial and generalized NCSE are the two main subtypes, but they can be poorly distinguished on EEG. 3. Diagnosis of NCSE relies upon clinical testing, empiric benzodiazepine treatment, and clear documentation of the patient’s best response or state before and after treatment. REFERENCES 1. Craven W, Faught E, Kuzniecky R, et al: Residual electrographic status epilepticus after control of overt electrical seizures [abstract]. Epilepsia 1995; 36:S46. 2. Fountain NB, Lothman EW: Pathophysiology of status epilepticus. J Clin Neurophysiol 1995; 12(4):326–342. 3. Granner MA, Lee SI: Nonconvulsive status epilepticus: EEG analysis in a large series. Epilepsia 1994; 35(1):42–47. 4. Kaplan PW: Nonconvulsive status epilepticus in the emergency room. Epilepsia 1996; 37(7):643–650. 5. Kirby D, Fountain NB, Quigg M: Standardized mental status testing for nonconvulsive status epilepticus. Am J Electroneurodiagnostic Technol 2004; 44(3):199–201. 6. Krumholz A, Sung GY, Fisher RS, et al: Complex partial status epilepticus accompanied by serious morbidity mortality. Neurology 1995; 45(8):1499–1504. 7. Quigg M, Shneker B, Domer P: Current practice in administration and clinical criteria of emergent EEG. J Clin Neurophysiol 2001; 18(2):162–165. 8. Scholtes FB, Renier WG, Meinardi H: Non-convulsive status epilepticus: Causes, treatment, outcome in 65 patients. J Neurosurg Neurol Psychiatry 1994; 61:93–95.

Status Epilepticus

261

9. Treiman DM, Meyers PD, Walton NY, et al: A comparison of four treatments for generalized convulsive status epilepticus. Veterans Affairs Status Epilepticus Cooperative Study Group. N Engl J Med 1998; 339:792–798. 10. Young GB, Jordan KG, Doig GS: An assessment of nonconvulsive seizures in the intensive care unit using continuous EEG monitoring: An investigation of variables associated with mortality. Neurology 1996; 47(1):83–89.

262

Status Epilepticus

PATIENT 95 A 60-year-old woman with decline in mental status

An EEG is requested to determine the etiology of decline in the level of consciousness in a 60-year-old woman with a history of lung transplant. She has no laboratory evidence of toxic-metabolic encephalopathy. No convulsive seizures are witnessed, but persistent leftward eye deviation and myoclonus are present. She is taking immunosuppressive and antihypertensive medications. The study demonstrates nearly continuous runs of 2–3 cps triphasic waves of varying morphology that are often generalized but sometimes have higher amplitudes in the vertex and left parasagittal regions. External stimulation does not change the patterns, and there is no detectable anteriorposterior lag among triphasic waves. Two samples are shown, one before administration of 5 mg IV diazepam (‘‘before diazepam’’) and another about 1 minute after administration (‘‘after diazepam’’). All clinical movements cease during the second sample. Question:

What is the diagnosis?

Status Epilepticus

263

Answer: Clinical and electrographic NCSE marked by clinical improvement coinciding with resolution of 2.75–3 cps diffusely distributed sharp-wave/slow-wave complexes with a triphasic morphology. Discussion: The ictal discharges present in The diagnostic criteria for ictal discharges in NCSE can be easy to confuse with other rhythmic NCSE are controversial, but some guidelines exactivities that do not represent seizure activity. ist. Proposed primary and secondary criteria are Triphasic waves of toxic-metabolic or ictal ori- as follows: gins may be indistinguishable from each other.

Primary Criteria 1. Repetitive generalized or focal spikes, sharp waves, spike-and-wave, or sharp-and-slow wave complexes at more than 3 per second. 2. Repetitive generalized or focal spikes, sharp waves, spike-and-wave, or sharp-and-slow wave complexes at fewer than 3 per second and secondary criterion no. 4. 3. Sequential rhythmic waves and secondary criteria 1, 2, and 3 with or without secondary criterion no. 4.

Secondary Criteria 1. 2. 3. 4.

264

Incrementing onset: increase in voltage or increase or slowing of frequency. Decrementing offset: decrease in voltage or frequency. Post-discharge slowing or voltage attenuation. Significant improvement in clinical state or baseline EEG after intravenous antiepileptic drug (AED). Status Epilepticus

Of course, the range of potential ictal discharges allowed by such flexible criteria is vast. In this patient’s case, rhythmic sharp and slow-wave complexes (with a triphasic morphology) with a frequency just under 3 cps (primary criterion no. 2) attenuated with benzodiazepines (secondary criterion no. 4). This tracing also demonstrated rhythmic evolution of triphasic waves as the EEG progressed, fulfilling secondary criteria no.1 and no. 2, but this finding is not discernible in these brief samples. Note that attenuation of rhythmic activity with benzodiazepines is not proof of NCSE, just supportive of the diagnosis. Patients with triphasic waves of metabolic-toxic origin can also show resolution of triphasic waves with benzodiazepines, supposedly on the basis of transiently worsened encephalopathy from sedation. Clinical examination, therefore, remains an important component in the use of benzodiazepines in the diagnosis of NCSE. Only with clinical improvement in correlation with electrographic resolution is there a definitive diagnosis of NCSE.

Clinical Pearl Rhythmic discharges of ictal origin must be distinguished from those of nonictal origin on the basis of morphology, evolution, reactivity, and clinical and electrographical responses to anticonvulsant medications. REFERENCES 1. Chong DJ, Hirschl J: Which EEG patterns warrant treatment in the clinically ill? Reviewing the evidence for treatment of periodic epileptiform discharges and related patterns. J Clin Neurophysiol 2005; 22:79–91. 2. Fountain NB, Waldman WA: Effects of benzodiazepines on triphasic waves: Implications for nonconvulsive status epilepticus. J Clin Neurophysiol 2001; 18(4):345–352. 3. Young GB, Jordan KG, Doig GS: An assessment of nonconvulsive seizures in the intensive care unit using continuous EEG monitoring: An investigation of variables associated with mortality. Neurology 1996; 47(1):83–89.

Status Epilepticus

265

PATIENT 96 A 74-year-old man in a coma after cardiac arrest

An EEG is requested to evaluate possible NCSE in a patient who has two convulsive seizures and does not clinically improve in level of consciousness 4 hours after the last event. Seizures occur after cardiac arrest, resuscitation, and acute myocardial infarction. His medications include phenytoin and lorazepam. The baseline EEG is recorded at bedside with patient comatose and ventilated (‘‘before diazepam’’). ‘‘Arms extended’’ in the second sample (‘‘after diazepam’’) corresponds to decerebrate posturing evoked by painful stimulation, a response not present during the baseline recording. Questions: NCSE?

266

What are the baseline findings? Does the second sample confirm a diagnosis of

Status Epilepticus

Answers: PLEDs with a periodicity of approximately 1 per second appear across the right hemisphere on suppressed background activities. Clinical improvement and electrographic resolution following administration of benzodiazepines confirm a diagnosis of NCSE. Discussion: In the paradigm provided by animal models of status epilepticus, PEDs represent the late effect of chronic, untreated CSE. As CSE progresses, discrete clinical seizures merge to form waxing and waning obtundation. Bursts of ictal activity, originally separated by abnormal background activity, progress to nearly continuous rhythmic ictal discharges. If allowed to continue, clinical seizure activity may evolve to stupor or coma without clear convulsions. The electrographic end stage in this process is PEDs. On the other hand, PEDS occur late in experimental status epilepticus. PEDS may represent electrographic evidence of acute neuronal injury following prolonged seizures. Thus, PEDs may represent the injury resulting from long-lasting seizures. Subsequent findings in humans with severe status epilepticus, however, document that PEDs resolve with successful anticonvulsant therapy, providing evidence that PEDs are ictal. Clinical and experimental work have converged to suggest that PEDs in a comatose patient should lead to the diagnosis of status epilepticus, even if clinical movements are absent or are extremely subtle. In this light, NCSE is a potentially treatable cause of coma that requires an

Status Epilepticus

emergent EEG to make the diagnosis. The main problem facing clinicians is that no clear markers in the EEG itself, divorced from clinical information, reliably separate PEDs of ictal origin from those that designate acute, destructive lesions. PEDs of ictal origin may show evidence of spatial and temporal evolution: Amplitude or frequency may wax and wane, or spontaneously stop only to recur, and PEDs may show variable location. PEDs of ictal origin should attenuate with a trial of anticonvulsant medication. PEDs from acute structural lesions, on the other hand, remain continuous, poorly reactive, localized to one distribution, and do not attenuate with anticonvulsant medication. In this patient’s case, witnessed convulsions evolved to a continuous state of coma and NCSE. Up to 20% of convulsive status epilepticus continues to have NCSE following resolution of clinical movements. Residual NCSE following convulsive status epilepticus is a poor prognostic sign, with 65% dying within 30 days of presentation. In comparison, only 27% of patients died whose CSE resolved without evidence of NCSE. An EEG is recommended to determine the electrographic success of treatment of status

267

epilepticus if the patient shows no evidence of activity with anticonvulsant medication, the clinical recovery following treatment. Despite the patient did not recover and died 2 days after this successful treatment of electrographic seizure recording.

Clinical Pearls 1. PEDs may represent ongoing electrographic seizure activity in severe seizures and encephalopathies. 2. Distinguishing ictal from nonictal PEDs requires clinical examination and adjunctive anticonvulsant use during the EEG. 3. An EEG should be done following resolution of CSE, if there is no clinical improvement to rule out ongoing NCSE. REFERENCES 1. Chong DJ, Hirschl J: Which EEG patterns warrant treatment in the clinically ill? Reviewing the evidence for treatment of periodic epileptiform discharges and related patterns. J Clin Neurophysiol 2005; 22:79–91. 2. Craven W, Faught E, Kuzniecky R, et al: Residual electrographic status epilepticus after control of overt electrical seizures [abstract]. Epilepsia 1995; 36:S46. 3. Quigg M, Schneker B, Domer P: Current practice administration of emergent EEG. J Clin Neurophysiol 2001; 18:162–165. 4. Treiman DM, Meyers PD, Walton NY, et al: A comparison of four treatments for generalized convulsive status epilepticus. Veterans Affairs Status Epilepticus Cooperative Study Group. N Engl J Med 1998; 339:792–798. 5. Treiman DM, Walton NY, Kendrick C: A progressive sequence of electroencephalographic changes during generalized convulsive status epilepticus. Epilepsy Res 1990; 5(1):49–60.

268

Status Epilepticus

PATIENT 97 A 71-year-old man with stupor and asterixis

A 71-year-old man with hepatitis presents with worsening confusion and negative myoclonus (asterixis). An EEG is requested to evaluate possible NCSE. He takes no CNS-active medication. The recording is performed with the patient lethargic and intermittently trembling. The first sample (‘‘before diazepam’’) shows 20 seconds during which external stimulation is performed. The second 10-second sample (‘‘during injection’’) shows the patient during unstimulated rest during administration of intravenous diazepam. The third sample (‘‘after diazepam’’) shows activities 20 minutes after the diazepam is flushed. The patient shows no clinical changes throughout the study. Questions: What are the patterns in the first and second samples? What are the activities in the third sample? Are the findings compatible with NCSE?

Status Epilepticus

269

Answers: The first and second samples show PEDs occurring at a frequency of 1–2 cps on background activities of low-amplitude theta activities. PEDs often have a triphasic morphology and sometimes occur rhythmically with loss of background activities. During the first sample, activities attenuate with external stimulation. In the third sample, PEDs have largely resolved following administration of diazepam. Despite the effect of diazepam on periodic activities, the finding that they resolve with stimulation suggests that PEDs/triphasic waves arise from metabolic or toxic encephalopathies rather than ongoing NCSE. Discussion: One procedure to help the diagnosis of NCSE is administration of anticonvulsants during the EEG. Resolution of rhythmic EEG activity supports, but does not prove, a diagnosis of NCSE. The reason for this cautious interpretation is that benzodiazepines attenuate triphasic waves of toxic-metabolic origin by further decreasing level of consciousness. The only time benzodiazepine administration is diagnostic in NCSE is when

clinical improvement accompanies resolution of the rhythmic discharge on EEG. In this patient’s case, external stimulation attenuates triphasic waves (sample ‘‘before diazepam’’), suggesting that rhythmic activity is reactive and unlikely to be ictal. The physician decided to administer diazepam, despite the findings of reactivity because attenuation was not a consistent response.

Clinical Pearls 1. Diagnosis of NCSE requires evaluation of all of the findings at hand, including patient responses to reactivity and to diazepam. 2. Administration of benzodiazepines may resolve triphasic waves by inducing a transient exacerbation in encephalopathy. REFERENCE 1. Chong DJ, Hirschl J: Which EEG patterns warrant treatment in the clinically ill? Reviewing the evidence for treatment of periodic epileptiform discharges and related patterns. J Clin Neurophysiol 2005; 22:79–91. 2. Fountain NB, Waldman WA: Effects of benzodiazepines on triphasic waves: Implications for nonconvulsive status epilepticus. J Clin Neurophysiol 2001; 18(4):345–352.

270

Status Epilepticus

PATIENT 98 An 8-year-old boy with static encephalopathy and recurrent myoclonus

An 8-year-old boy presents with spells of bilateral arm and trunk jerking and no clear alteration of consciousness. He has severe static encephalopathy after viral meningoencephalitis and is blind and quadriplegic. He takes no CNS-active medications. The EEG is obtained with the patient awake and with continuous, repetitive jerks of the trunk. Question:

What is the origin of periodic discharges in this sample?

Status Epilepticus

271

Answer: Periodic bursts of diffuse muscle activity occur upon low-amplitude, suppressed background activities. Discussion: Myoclonus can have cortical, subcortical, or spinal localizations. Cortical myoclonus is sometimes called epileptic myoclonus and is accompanied by evidence of epileptiform discharges. Subcortical or spinal myoclonus (also known as nonepileptic myoclonus) often has no evidence of epileptic activity. Special recording techniques that record samples of EEG and are time-locked to episodes of myoclonus can confirm the sequential order of epileptic potentials to myoclonic jerks. With the use of time-locked recording and sample-averaging (a technique that increases

signal strength by averaging reproducible signal over random noise), the timing of myoclonus versus cortical activity can be quantified. Responses to anticonvulsants in myoclonus can be unpredictable because clinical symptoms can respond but electrographic discharges may not. In this patient’s case, the bursts of electrical activity consist of muscle and movement artifact. Such findings often support a diagnosis of nonepileptic myoclonus, but the present tracing is ambiguous because of the masking of possible cortical potentials by artifact.

Clinical Pearls 1. Myoclonus may be classified by routine EEG into epileptic and nonepileptic myoclonus. 2. Persistent movement or dystonia may mask EEG findings with artifact. REFERENCE 1. Niedermeyer E, Riley T: Myoclonus the electroencephalogram: A review [112 refs]. Clin Electroencephalogr 1979; 10(2):75–95. 2. Shibasaki H, Yamashita Y, Tobimatsu S, Neshige R: Electroencephalographic correlates of myoclonus. Adv Neurol 1986; 43:357–372.

272

Status Epilepticus

PATIENT 99 A 45-year-old man with a coma and myoclonus

A 45-year-old man presents with a coma and myoclonus. He has hepatitis C and alcoholic cirrhosis with recent pneumonia, sepsis, and upper gastrointestinal (GI) hemorrhage. Rhythmic jerking begins on the second day of coma. Listed medications include lorazepam drip. The EEG is obtained while the patient is sedated, intubated, and comatose. Continuous generalized jerking is present at baseline (‘‘periodic myoclonus’’) and is mostly absent (except for residual facial clonus) after administration of vecuronium (‘‘after vecuronium’’). Questions: From interpretation of the baseline sample only, what are the possible sources of the bursts? What does administration of vecuronium accomplish in the interpretation? Is there evidence of status epilepticus?

Status Epilepticus

273

Answers: Quasi-periodic bursts of extremely sharp activities, in the first sample, can be either bursts of muscle artifact or PEDs. Vecuronium, a paralytic agent, resolves all-but-residual myoclonus. The EEG responds similarly, showing near resolution of bursts (later on, all muscular and apparent EEG activity resolves, but worsening electrical noise prevents its presentation here). Paralysis confirms that quasi-periodic bursts are artifact, leaving low-amplitude, diffusely distributed delta activity. The diagnosis is nonepileptic myoclonus. Discussion: Vecuronium and other paralytic agents can aid interpretation of the EEG by removing the obscuring effect of muscle artifact. Of course, paralytic agents can be administered only in comatose or thoroughly sedated subjects to prevent the possibility of conscious paralysis, a potentially terrifying experience. Paralytic agents or, more precisely, those who use paralytic agents, have earned poor reputations with those charged in diagnosis of status epilepticus. Paralytic agents are often used emergently to aid in the establishment and maintenance of

the airway in violently convulsing patients. Occasionally, however, one may use paralytic agents to ‘‘treat’’ persisting convulsive movements, forgetting that electrical seizure activity may persist despite neuromuscular blockade. Iatrogenic paralysis induced after witnessed seizure activity is a clear indication for emergent EEG to ensure against iatrogenic NCSE. In this patient’s case, no cerebral activities coincide with myoclonus, best classifying the syndrome as nonepileptic myoclonus.

Clinical Pearls 1. Paralytic agents may allow interpretation of EEG in tracings obscured by muscle artifact in appropriate patients. 2. Iatrogenic paralysis following a witnessed seizure requires a follow-up EEG to confirm resolution of electrical seizure activity. REFERENCE 1. Quigg M, Shneker B, Domer P: Current practice in administration and clinical criteria of emergent EEG. J Clin Neurophysiol 2001; 18(2):162–165.

274

Status Epilepticus

PATIENT 100 A 74-year-old man with a coma and myoclonus following cardiac arrest

An EEG to aid in ongoing treatment of CSE is requested for a 74-year-old man after cardiac resuscitation for asystolic arrest. He has generalized convulsive seizures, confirmed by witnesses, immediately after resuscitation. After emergent treatment with lorazepam, he continues to have myoclonic jerks. The EEG is recorded with the patient comatose and jerking. An EMG channel is placed on his right hand. He remains on phenytoin and cardiopressors. Lorazepam and other sedatives are discontinued 5 hours before the recording. Questions: Is vecuronium required to interpret this study? Is it status epilepticus? What prognostic significance does this pattern have?

Status Epilepticus

275

Answers: The recording shows quasi-periodic bursts of multiple spikes occurring on suppressed background activities. The EMG channel shows that multiple spikes precede arm jerks by approximately 200 milliseconds. Because cortical activity precedes EMG changes, vecuronium is not necessary because EMG does not obscure cortical activities. This pattern is consistent with epileptic myoclonus. Its continuous nature is consistent with status myoclonus or myoclonic status epilepticus. It is associated with subsequent death or severe neurologic impairment. Discussion: Myoclonic status epilepticus is defined as continuous myoclonus, in addition to other epileptic seizures that persist for over 30 minutes. The myoclonic seizures can be cortical in origin or arise from subcortical or reticular structures, and both can occur together. Some authors distinguish myoclonic status epilepticus from status myoclonus, in which myoclonus, of either epileptic or nonepileptic origin, is not accompanied by other seizure types. Myoclonic status epilepticus occurs most frequently after cardiac arrest and cerebral anoxia. Depending on the differences in definition and inclusion in various studies, myoclonic status epilepticus occurs after 3–37% of cardiac resuscitations. Out of a combined total of 232 patients studied in the references that follow, myoclonic status epilepticus is observed in 54 (23%). All 54 died subsequent to discovery of myoclonic status epilepticus, despite anticonvulsant therapy. The poor response to anticonvulsant therapy, both in the ability to stop myoclonus and in the

lack of effect in outcome, raises the possibility that, in the spectrum of PEDs versus ictal discharges, myoclonic status epilepticus is an agonal rhythm of diffuse, acute, and severe neuronal injury rather than a seizure state. To withhold anticonvulsant therapy on this interpretation remains controversial. Many physicians continue to treat patients with evidence of cortical myoclonus or mixed seizures, in addition to myoclonus. In this patient’s case, the placement of EMG electrodes on a limb allowed classification of this pattern as an epileptic myoclonus, because cortical discharges preceded and, presumptively, caused myoclonic jerks. If myoclonic jerks occurred simultaneously with cortical discharges, the classification into epileptic (cortical) versus nonepileptic (subcortical) would remain ambiguous, because the short duration, polyphasic spike activity could alternatively be interpreted as muscle artifacts rather than cortical activity. In the latter situation, temporary paralysis could aid in interpretation.

Clinical Pearls 1. Myoclonic status epilepticus is myoclonus and other seizure types persisting beyond 30 minutes. 2. Myoclonic status epilepticus portends severe neuronal injury and death in most subjects if seen after cerebral anoxia. REFERENCES 1. Krumholz A, Weiss HD: Outcome from coma after cardiopulmonary resuscitation: Relation to seizures myoclonus. Neurology 1988; 38:401–405. 2. Wijdicks EF, Parisi JE, Sharbrough FW: Prognostic value of myoclonus status in comatose survivors of cardiac arrest. Ann Neurol 1994; 35(2):239–243. 3. Young GB, Gilbert JJ, Zochodne DW: The significance of myoclonic status epilepticus in postanoxic coma. Neurology 1990; 40:1843–1848.

276

Status Epilepticus

INDEX Page numbers followed by f denote figures. A Abnormal posterior slowing, 54 Absence, 149 Absence seizures, 74, 146 Absence status epilepticus, 260 Absent brainstem reflexes, 247–248 AC. See Alternating current AC circuit, 3 Acquired aphasia, 176–177 Activation procedure, 71 Active sleep, 96, 109, 193 Activte moyenne, 109 ADA. See Arrhythmic delta activity AED. See Antiepileptic drug Aliasing, 13 Alpha coma, 246 Alpha rhythm, 48 asymmetry of, 60 first stages of, 52 frequency of, 48 location of, 48 morphology of, 48 paradoxical, 219 reactivity of, 48 slow, 62 slowing/attenuation of, 80 state-dependence of, 48 symmetry of, 48 Alternating current (AC), 3 Amp, 1 Amplifier, 15–16 Amplitude, 25 of encephalopathy, 216–217 of Mu rhythm, 64 Analog-to-digital (ATD), 13 Anterior dysrhythmia, 104 Anteriorly dominant rhythmic delta activity, 229 Antiepileptic drug (AED), 264 Aortic aneurysm dissection, 243–244 Aphasia, 176, 252–253 acquired, 176–177 epileptical, 177 Apnea, 113–115, 190–193 during active sleep, 193 brief, 193 central, 191 epileptic, 191 idiopathic, 191 mixed, 191 neonatal nonepileptic, 191 obstructive, 191 Arrhythmic activity, 39 Arrhythmic delta activity (ADA), 202 Artifactual attenuation, 240 Astatic seizures, 164 ATD. See Analog-to-digital

Atonic seizures, 197 Atypical absence seizures, 195 Atypical febrile convulsion, 175 Auras, 180, 207–208 Autism, 65–67 Averaged ear inputs (AVG), 241 AVG. See Averaged ear inputs Axial spasms, 197 B Basic electricity, 1–2 BECTS. See Benign childhood epilepsy with centrotemporal spikes Benign childhood epilepsy with centrotemporal spikes (BECTS), 120 Benign epileptiform transients, 131 Benign occipital epilepsy of childhood (BOEC), 123, 124 Benign Rolandic epilepsy (BRE), 120 Benzodiazepines, 270 Bilateral limb extension, 159–160 Bilateral limb flexion, 161 Bimetallic artifact, 18 Biocalibration, 33 Bipolar longitudinal montages, 21 Bipolar montages, 21, 27 Bipolar transverse montages, 21 Bit depth, 13 Blocking, 11 BOEC. See Benign occipital epilepsy of childhood Brainstem reflexes, 247–248 BRE. See Benign Rolandic epilepsy Breach rhythm, 66, 214, 238 Breath-holding events, 199 Brief apneas, 193 Burst-suppression, 244 C CAE, 146 Calibration requirements, 32–33 Capacitance, 3 Capacitive reactance (Xc), 4 Capacitor, 3, 7 Carbamazepine, 61, 122 Cardiac arrest, 241–242, 245–246, 266–268, 275–276 CDE. See Cerebral death exam Central apnea, 191 Central nervous system (CNS), 74 Cerebral death exam (CDE), 248 Cerebral palsy, 140–141 Cerebrospinal fluid (CSF), 253 Channels, 21–23 eye lead, 34–35 isopotential, 28 Charge, 1 Childhood absence epilepsy (CAE), 146

277

Chronic obstructive pulmonary disease (COPD), 254 Circuits AC, 3 DC, 3 elementary, 3–4 CJD. See Creutzfeldt-Jakob disease Clipping, 11 Clonidine, 68 CMRR. See Common mode rejection ratio CNS. See Central nervous system (CNS) Coma, 241–242, 245–246, 273–274, 275–276 Common mode rejection, 15 Common mode rejection ratio (CMRR), 15 Complex convulsion, 175 Complex partial seizures, 194–195, 205–206 Computed tomography (CT), 202 Confusion, 75–77, 89–90, 235–236, 252–253, 256–257 Consciousness, 57–58 Contaminated reference, 20 Continuous activities, 96, 109 Continuous spike-wave of slow wave sleep, 177 Convulsion(s) atypical febrile, 175 complex, 175 single febrile, 175 syncopal, 199 tonic-clonic, 150–151 Convulsive status epilepticus (CSE), 260 COPD. See Chronic obstructive pulmonary disease Cortical myoclonus, 272 Corticectomy, 213–214 Coulomb, 1 Creutzfeldt-Jakob disease (CJD), 250, 255 CSF. See Cerebrospinal fluid CT. See Computed tomography Ctenoid, 135 Current, 1 AC, 3 DC, 3 Cutoff frequency, 9–10 D DC. See Direct current DC circuit, 3 Decibels, 15 Deep sleep, 84 Delta absence, 149 Delta brushes, 102 Depression, 55–56, 61–62, 87–88, 220–221 major, 142–143 Developmental delay, 163 Diaphoresis/unresponsiveness, 70–72 Diazepam, 259 Differential amplifier, 15–16 Diffuse encephalopathy altered mental state (70-year-old man), 215–217 aortic aneurysm dissection (75-year-old man), 243–244 cardiac arrest/coma (68-year-old man), 245–246 depression/parkinsonism (77-year-old woman), 220–221

278

Diffuse encephalopathy (Continued) EEG effected by, 215–248 electroconvulsive therapy (53-year-old woman), 218–219 fulminant encephalitis/absent brainstem reflexes (23-year-old man), 247–248 lethargy (5-year-old girl), 230–231 lethargy (59-year-old woman), 222–223 postoperative confusion (61-year-old man), 235–236 pseudotumor cerebri/intermittent lethargy (49-year-old woman), 228–229 psychosis (55-year-old man), 226–227 spells (79-year-old woman), 224–225 stupor (71-year-old woman), 237–238 stupor/jaundice (85-year-old woman), 232–233 Digital EEG, 13 Dipole, 30 Direct current (DC), 3 Discharges. See also Generalized discharges; Ictal discharges; Interictal epileptiform dischargers generalized polyspike-wave, 155 Discontinuous activities, 96 Discrete lesions, 210 Distribution, 37 Drop attacks, 163, 196–197 Drowsiness, 56 hypersynchrony of, 94 RMTD, 131 Dysmaturity, 96, 117 E Ear electrodes, 20 Early infantile epileptic encephalopathy (EIEE), 160 Early neonatal myoclonic encephalopathy (ENME), 160 Earth ground, 5 ECA. See Estimated conceptional age ECS. See Electrocerebral silence ECT. See Electroconvulsive therapy EEG. See Electroencephalogram EEG machines cutoff frequency and, 9 sensitivity and, 11f EEG tracing, 37 EKG. See Electrocardiogram EKG artifact, 45 Electrical field, 24 Electrical ground, 1 Electrical potential, 1 Electrical safety, 5–6 Electrical status epilepticus of sleep (ESES), 177 Electricity, 1–2 Electrocardiogram (EKG), 5 Electrocerebral silence (ECS), 114, 217 Electroconvulsive therapy (ECT), 55, 87, 218–219 Electrode(s) 10-20 system for, 19f ear, 20 EEG, 17 electroencephalographic, 17–18

Index

Electrode(s) (Continued) nasopharyngeal, 20 placement of, 18, 19f scalp, 17f silver-chloride, 17 true temporal, 20 Electrode placement, 248 Electrode pop, 90 Electrodecremental seizure, 162 Electroencephalogram (EEG), 1. See also Convulsion(s); Epilepsy; Focal sharp transients; Normal electroencephalogram of sleep; Normal waking electroencephalogram; Seizure(s) acquisition of, 17–35 alpha rhythm and, 48, 49 calibration/technical requirements of, 32–33 diffuse encephalopathy effects on, 215–248 altered mental state (70-year-old man), 215–217 aortic aneurysm dissection (75-year-old man), 243–244 cardiac arrest/coma (68-year-old man), 245–246 coma after cardiac arrest (34-year-old woman), 241–242 depression/parkinsonism (77-year-old woman), 220–221 electroconvulsive therapy (53-year-old woman), 218–219 fulminant encephalitis/absent brainstem reflexes (23-year-old man), 247–248 lethargy (5-year-old girl), 230–231 lethargy (59-year-old woman), 222–223 motor vehicle accident/head trauma (16-year-old girl), 239–240 postoperative confusion (61-year-old man), 235–236 pseudotumor cerebri/intermittent lethargy (49-year-old woman), 228–229 psychosis (55-year-old man), 226–227 spells (79-year-old woman), 224–225 stupor (71-year-old woman), 237–238 stupor/jaundice (85-year-old woman), 232–233 digital, 13 EKG artifact/cerebral activity and, 45 encephalopathy correlation with, 236 filters for, 7–8 intensive video, 184–185 local lesions on, 201–202 bizarre behavior (6-year-old boy), 209–210 complex partial seizures (30-year-old man), 205–206 intractable epilepsy after corticectomy (55-year-old woman), 213–214 left hemiparesis (73-year-old woman), 201–202 left intracranial hemorrhage (28-year-old woman), 211–212 MTLE (33-year-old man), 203–204 physic auras (28 year-old woman), 207–208 localization and, 24 neonatal apnea (term infant), 113–115 hypotonia (premature infant), 101–102

Index

Electroencephalogram (EEG) (Continued) interpretation of, 99 jitteriness (term infant), 103–110 spells (6-week-old infant), 111–112 normal, of sleep, 79–94 depression with psychotic features (30-year-old woman), 87–88 inattentive spells (7-year-old boy), 93–94 intermittent confusion (22-year-old man), 89–90 overnight video-EEG recording (23-year-old woman), 79–86 seizure concerns (7-month-old infant), 91–92 normal waking, 47–77 autism/episodic rage attacks (17-year-old boy), 65–67 depression (40-year old man), 55–56 depression/spells (36-year-old woman), 61–62 head trauma/episodic rage attacks (year-old boy), 68–69 headaches (20-year-old man), 47–49 headaches and inattention (4-year-old girl), 73–74 headaches/confusion (14-year-old girl), 75–77 new-onset seizures (21-year-old man), 59–60 possible absence seizures (16-year-old boy), 63–64 seizures after urinary tract infection (6-year-old girl), 53–54 spells of diaphoresis/unresponsiveness (51-year-old woman), 70–72 spells of loss of consciousness (25-year-old man), 57–58 staring spells (12-month-old boy), 50–52 pattern of deflections of, 24–31 physiologic, patterns, 96 prognostic value of EEG, 236 shark-fin appearance of output waveform from, 32f Electroencephalographers, 39 arrhythmic activity of, 39 background activity in, 39 evolution of, 39 periodic activity of, 39 rhythm of, 39 Electroencephalographic electrodes, 17–18 Electroencephalographic findings, 37–45 Electromyographic (EMG), 69 Electro-oculograph (EOG), 34, 108–109 Elementary circuits, 3–4 EMG. See Electromyographic Encephalitis, 116–118, 247–248 Encephalopathy, 216. See also Diffuse encephalopathy EIEE, 160 ENME, 160 Encoches frontales, 104, 105 End-of-chain, 28 Energy, 2 Enhanced beta activity, 66 ENME. See Early neonatal myoclonic encephalopathy Enucleated right eye, 206 EOG. See Electro-oculograph

279

Epilepsy, 131, 132–133, 160. See also Generalized epilepsy; Interictal epileptiform dischargers (IED); Localization-related epilepsy; Status epilepticus AED and, 264 BECTS, 120 benign Rolandic, 120 BOEC, 123, 124 BRE, 120 CAE, 146 childhood absence, 146 frontal lobe, 153 intractable, 213–214 JME, 155 limbic, 128 localization-related, 119–143 MTLE, 128–129, 179–180, 184–185, 203–204 myoclonic, 165–166 myoclonic-astatic, 197 PME, 166 status epilepticus and, 259–261 Epileptic apnea, 191 Epileptic encephalopathy, 160 Epileptic seizures, 179–199 apnea (1-week-old female infant), 192–193 apnea (1-week-old infant), 190–191 complex partial seizures (37-year-old man), 194–195 drop attacks (17-month-old girl), 198–199 face-twitching/tremulousness (39-year-old man), 182–183 intensive video-EEG (53-year old man), 184–185 Lennox-Gastaut syndrome /drop attacks, 196–197 mesial temporal lobe epilepsy/right occipital cystic lesion (24-year-old man), 179–180 right body clonus (2-day old term infant), 188–189 spells (30-year old woman), 186–187 Epileptic syndrome, 161 Epileptical aphasia, 177 Epilepticus. See also Status epilepticus absence status, 260 CSE, 260 ESES, 177 myoclonic status, 276 NCSE, 260 partial status, 260 Epileptiform, 41 Epileptiform morphology, 118 Episodic tinnitus, 136 ESES. See Electrical status epilepticus of sleep Estimated conceptional age (ECA), 95 Evolution, 39f Eye lead channels, 34–35, 35 Eye leads, 34 Eye movement, 35, 227 F Face twitching, 182–183 Fast alpha rhythm variant, 58 Fast frequency, 66

280

Febrile convulsion atypical, 175 complex, 175 single, 175 3-year old boy, 174–175 Fever, 252–253 Filter cutoff frequency, 10 Filters, 7–8 application of different, to same signal, 9f cutoff frequency and, 9–10, 10 HFF, 7, 8 LFF, 7, 8 notch, 7 understanding, 7 FIRDA. See Frontally dominant intermittent rhythmic delta activity FLAIR. See Fluid-attenuation inversion recovery Fluid-attenuation inversion recovery (FLAIR), 128 Fluoxetine, 68 Focal arrhythmic delta activity, 238 Focal discharge, 28 Focal enhancement, 214 Focal sharp transients, 119–143 episodic tinnitus (35-year-old woman), 136 generalized seizures (32-month-old boy), 138–139 major depression with psychotic features (30-year-old woman), 142–143 medically intractable complex partial seizures (54-year-old woman), 127–129 nocturnal hemiconvulsions/visual seizures (11-year-old boy), 122–124 nocturnal seizures (7-year-old boy), 119–121 occipital sharp transients during sleep (21-year-old woman), 125–126 paroxysmal paresthesia (32-year-old woman), 134–135 seizures resumption (40-year-old epileptic man), 132–133 spells/cerebral palsy (4-year-old girl), 140–141 visual hallucinations (56-year-old drowsy man), 130–131 Focal slowing, 202 Frequency, 3, 37 of alpha rhythm, 48 of CAE, 146 of encephalopathy, 216 of enhance beta activity, 66 fast, 66 filter cutoff, 10 of generalized poly-spike discharges, 155 mixed, 109 of mu rhythm, 64 of PEDs, 250 of phantom spike-waves, 151 of RMTD, 131 sampling, 13 Frontal lobe epilepsy, 153 Frontal slowing, 227 Frontally dominant intermittent rhythmic delta activity (FIRDA), 223, 229

Index

Fulminant encephalitis, 247–248 Fuzziness, 10 G Gain, 15, 16 Generalized convulsions, 176–177 Generalized discharges, 145–168 congenital abnormalities/bilateral limb flexion (8-month-old male infant), 161–162 developmental delay/drop attacks (3-year-old girl), 163 light provoked seizures/morning myoclonic seizures (12-year-old girl), 154–155 medically intractable generalized seizures (29-year old man), 152–153 metachromatic leukodystrophy, 167–168 photic discomfort (14-year-old girl), 157–158 progressive neurologic deterioration/myoclonic epilepsy (4-year-old boy), 165–166 spells of bilateral limb extension/trunk flexion (12-week-old female infant), 159–160 staring spells (five-year old girl), 145–146 staring spells (7-year old girl), 148–149 tonic-clonic convulsions after motor vehicle accident (41-year-old woman), 150–151 Generalized epilepsy, 145–168, 163 progressive neurologic deterioration/myoclonic epilepsy (4-year-old boy), 165–166 spells of bilateral limb extension/trunk flexion (12-week-old female infant), 159 staring spells (five-year old girl), 145–146 Generalized polyspike-wave discharges, 155 Generalized seizures, 138–139 light provoked seizures/morning myoclonic seizures (twelve-year-old girl), 154–155 photic discomfort (14-year-old girl), 157–158 staring spells (7-year old girl with), 148–149 H Hallucinations, 130–131 Haloperidol, 87 Head trauma, 239 Headaches, 47–49, 73–77 Hematomas, 240 Hemiconvulsions, 122–124 Hemiparesis, 201–202 Herpes simplex viral encephalitis (HSVE), 253 Heteropias, 162 HFF. See High frequency filter High frequency filter (HFF), 7, 8 High voltage slow (HVS), 96, 107 High-amplitude responses, 76 HIHARS. See Hyperventilation-induced high-amplitude rhythmic slowing Hippocampal sclerosis, 128 Horizontal dipole, 30 HSVE. See Herpes simplex viral encephalitis HVS. See High voltage slow Hygromas, 240 Hypersynchrony, 94 hypnagogic, 94

Index

Hypersynchrony (Continued) hypnopompic, 94 Hypersynchrony of drowsiness, 94 Hyperventilation, 74, 148 Hyperventillation-induced high-amplitude rhythmic slowing (HIHARS), 149 Hypnagogic hypersynchrony, 94 Hypnopompic hypersynchrony, 94 Hypoxia, 114 Hypsarrhythmia, 162 Hyptonia, 101–102 I Iatrogenic paralysis, 274 Ictal discharges, 179–199 apnea (1-week-old female infant), 192–193 apnea (1-week-old infant), 190–191 complex partial seizures, 194–195 definition of, 180 drop attacks (17-month-old girl), 198–199 face-twitching/tremulousness (39-year-old man), 182–183 intensive video-EEG (53-year old man), 184–185 Lennox-Gastaut syndrome /drop attacks, 196–197 mesial temporal lobe epilepsy/right occipital cystic lesion (24-year-old man), 179–180 right body clonus (2-day old term infant), 188–189 spells (30-year old woman), 186–187 Ictal scalp recordings, 185 Idiopathic apnea, 191 IED. See Interictal epileptiform dischargers Impedance, 3, 4 Impedance mismatching, 15, 60 Inductance, 3 Infantile spasms, 162 Interburst intervals, 115 Interictal epileptiform dischargers (IED), 41–43, 170, 238 in children, 173 epileptical aphasia and, 177 sharp transient and, 43 single seizures and, 170 Intermittent confusion, 89–90 Intermittent photic stimulation, 76–77 Intermittent rhythmic delta activity (IRDA), 202 International 10–20 system, 19–20 Intracranial recordings, 180 Intractable epilepsy after corticectomy, 213–214 Intraventricular hemorrhage, 98 IRDA. See Intermittent rhythmic delta activity Isolated grounds, 5 Isopotential channels, 28 J Jaundice, 232–233 Jitteriness, 103–110 Juvenile myoclonic epilepsy (JME), 155 K K complexes, 82

281

L Lambda waves, 71, 72 Landau-Kleffner syndrome, 177 Lateral limb flexion, 161–162 Lateralis muscle spicules, 137 Left hemiparesis, 201–202 Lennox-Gastaut syndrome, 153, 164, 196–197 Lesions. See also Local lesions discrete, 210 local cortical, 240 occipital cystic, 179–180 Lethargy, 222–223, 228–231 LFF. See Low-frequency filter Light provoked seizures, 154–155 Light sleep, 90 Limb extension, 159 bilateral, 159–160 Limbic epilepsy, 128 Local cortical lesions, 240 Local lesions bizarre behavior (6-year-old boy), 209–210 complex partial seizures (30-year-old man), 205–206 intractable epilepsy after corticectomy (55-year-old woman), 213–214 left hemiparesis, 201–202 left intracranial hemorrhage (28-year-old woman), 211–212 MTLE (33-year-old man), 203–204 physic auras (28 year-old woman), 207–208 Localization, 24–31 of RMTD, 131 of small, sharp spikes, 133 Localization-related epilepsy, 119–143 major depression with psychotic features (30-year-old woman), 142–143 medically intractable complex partial seizures (54-year-old woman), 127–129 nocturnal hemiconvulsions/visual seizures (11-year-old boy), 122–124 nocturnal seizures (7-year-old boy), 119–121 occipital sharp transients during sleep (21-year-old woman), 125–126 paroxysmal paresthesia (32-year-old woman), 134–135 seizures resumption (40-year-old epileptic man), 132–133 visual hallucinations (56-year-old drowsy man), 130–131 Location, 37 of alpha rhythm, 48 of BECTS, 120 of BOEC, 123 of CAE, 146 of enhance beta activity, 66 of PEDs, 250 of phantom spike-waves, 151 of SREDA, 183 Low voltage irregular (LVI), 96, 109 Low-frequency filter (LFF), 7, 8 Lundborg disease, 166 LVI. See Low voltage irregular

282

M Magnetic resonance imaging (MRI), 202 Major depression, 142–143 Medically intractable complex partial seizures, 127–129 Medically intractable generalized seizures, 152–153 Memory loss, 254–255 Meningoencephalitis, 271 Mesial sclerosis, 128 Mesial temporal lobe epilepsy (MTLE), 128, 179–180, 184–185, 203–204 Metachromatic leukodystrophy, 167–168 Microvolts, 11 Mirtazapine, 87 Mixed apnea, 191 Modified hypsarrhythmia, 162 Montages, 21–23 bipolar, 21, 27 longitudinal, 21 transverse, 21 bipolar transverse, 21 referential, 21, 28f Morning myoclonic seizures, 154–155 Morphology, 37 of alpha rhythm, 48 of BECTS, 120 of CAE, 146 of encephalopathy, 216–217 epileptiform, 118 of generalized poly-spike discharges, 155 of mu rhythm, 64 of PEDs, 250 of phantom spike-waves, 151 of RMTD, 131 of small, sharp spikes, 133 spindle-form, 242 of SREDA, 183 of triphasic waves, 233 Motor vehicle accident, 239 Movement artifact, 221 MRI. See Magnetic resonance imaging MTLE. See Mesial temporal lobe epilepsy Mu rhythm, 64 Multifocal independent spikes, 167 Multiple heteropias, 162 Myoclonic epilepsy, 165–166 Myoclonic status epilepticus, 276 Myoclonic-astatic epilepsy, 197 Myoclonus, 273–274, 275–276 cortical, 272 nonepileptical, 272 photo, 76–77 recurrent, 271–272 spinal, 272 subcortical, 272 syncopal, 199 N Nasopharyngeal electrodes, 20 NCSE. See Nonconvulsive status epilepticus Neonatal nonepileptic apnea, 191

Index

Neonatal polygraphy, 95–118 apnea (term infant), 113–115 encephalitis (term infant), 116–118 hyptonia (premature infant), 101–102 intraventricular hemorrhage (premature infant), 98 jitteriness (term infant), 103–110 requirements of, 95 spells (6-week-old infant), 111–112 Neonatal seizures, 189 New-onset seizures, 59–60 Nocturnal hemiconvulsions, 122–124 Nocturnal seizures, 119–121 Nonconvulsive status epilepticus (NCSE), 217, 260, 261 clinical/electrographic NCSE, 264 diagnosis, 270 Nonepileptical myoclonus, 272 Normal EEG of sleep, 79–94 depression with psychotic features (30-year-old woman), 87–88 inattentive spells (7-year-old boy), 93–94 intermittent confusion (22-year-old man), 89–90 overnight video-EEG recording (23-year-old woman), 79–86 seizure concern of (7-month-old infant), 91–92 Normal waking EEG, 47–77 autism/episodic rage attacks (17-year-old boy), 65–67 consciousness loss (25-year-old man), 57–58 depression (40-year old man), 55–56 depression and spells (36-year-old woman), 61–62 diaphoresis/unresponsiveness (51-year-old woman), 70–72 year-old boy with head trauma/episodic rage attacks, 68–69 headaches (20-year-old man), 47–49 headaches/inattention (4-year-old girl), 73–74 new-onset seizures (21-year-old man), 59–60 possible absence seizures (16-year-old boy), 63–64 spells of headaches/confusion (14-year-old girl), 75–77 staring spells (12-month-old boy), 50–52 urinary tract infection (6-year-old girl), 53–54 Notch filter, 7 Nyquist’s theorem, 13 O Obstructive apnea, 191 Occipital cystic lesion, 179–180 Occipital intermittent rhythmic delta activity (OIRDA), 146, 231 Occipital sharp transients during sleep, 125–126 Ohm’s Law, 1–2 Ohtahara syndrome, 160 OIRDA. See Occipital intermittent rhythmic delta activity P Paper speed, 11–12 Paradoxical alpha rhythms, 219

Index

Paradoxical sleep, 86 Paralysis, 274 Paralytic agents, 274 Parkinsonism, 220–221 Paroxysmal paresthesia, 134–135 Partial status epilepticus, 260 Paxil, 61 PCR. See Polymerase chain reaction Peak-to-peak (p-p), 11 PEDs. See Periodic epileptiform discharges Pen rule, 21 Period, 3 Period discharges, 250 Periodic epileptiform discharges (PEDs), 217, 267–268 confusion/seizures (83-year-old woman), 256–257 fever/confusion/aphasia, 252–253 metastatic melanoma/obtundation (61-year-old woman), 249–251 rapidly progressive memory loss/somnolence (71-year-old man), 254–255 Periodic lateralized epileptiform discharges (PLEDs), 250, 253, 267–268 Petit mal variant, 164 Phantom spike-waves, 151 Phase, 3 Phase reversal, 24, 25 Phenobarbital, 114 Phenytoin, 59 Photic discomfort, 157–158 Photic driving, 210 photoelectric responses, 18 Photomyoclonus, 76–77 Photoparoxysmal response, 76–77, 155, 158 Physiologic EEG patterns, 96 PLEDs. See Periodic lateralized epileptiform discharges PME. See Progressive myoclonus epilepsies PNES. See Psychogenic non-epileptic pseudoseizures Polymerase chain reaction (PCR), 253 Polymicrogyria, 162 Polysomnography (PSG), 80 Positive occipital sharp transients of sleep (POSTS), 82, 126 Possible absence seizures, 63–64 Posterior low waves of youth, 54 Posterior rhythm, 57 Posterior slow waves of youth, 54 Posterior slowing, 54 Postoperative confusion, 235–236 POSTS. See Positive occipital sharp transients of sleep Potential, 1 p-p. See Peak-to-peak Progressive myoclonus epilepsies (PME), 166 Progressive neurologic deterioration, 165 Prolonged hypoxia, 114 Prolonged interburst intervals, 115 Pseudotumor cerebri, 228–229 PSG. See Polysomnography Psychogenic non-epileptic pseudoseizures (PNES), 187

283

Psychogenic non-epileptic seizures, 187 Psychomotor variant, 131 Psychosis, 226–227 Q Quiet sleep, 96 R Rapid Eye Movement (REM), 86 Rapidly progressive memory loss, 254–255 Reactivity, 37 of alpha rhythm, 48 of encephalopathy, 217 of mu rhythm, 64 of PEDs, 250 of phantom spike-waves, 151 Recurrent myoclonus, 271–272 Reference contamination, 143 Reference electrodes, 143 Referential montages, 21, 28f, 221 REM. See Rapid Eye Movement Remeron, 87 Resistance, 1, 4 Response testing, 195 Restlessness, 112 Rhythm. See also Alpha rhythm breach, 238 of electroencephalographers, 39 Rhythmic discharges, 265 Rhythmic midtemporal theta bursts of drowsiness (RMTD), 131 right body clonus, 188–189 RMTD. See Rhythmic midtemporal theta bursts of drowsiness Rolandic positive spikes, 117 Rolandic spikes, 121 S Salt bridge, 18 Sampling frequency, 13 Scalp edema, 18 Scalp electrodes, 17f, 18 Secondary bilateral synchrony, 153 Seizure(s), 53–54, 91–92, 116–118, 256–257. See also Epileptic seizures; Generalized seizures absence, 74, 146 astatic, 164 atonic, 197 atypical absence, 195 complex partial, 194–195, 205–206 electrodecremental, 162 epileptic, 179–199 generalized, 138–139 light provoked, 154–155 medically intractable complex partial, 127–129 medically intractable generalized, 152–153 morning myoclonic, 154–155 neonatal, 189 new-onset, 59–60 nocturnal, 119–121 possible absence, 63–64

284

Seizure(s) (Continued) psychogenic non-epileptic, 187 psychogenic non-epileptic pseudo, 187 resumption of, 132–133 simple partial, 180 single, 169–170 tonic, 197 visual, 122–124 Self-limited photoparoxysmal responses, 158 Sensitivity, 11–12 Sharp transient, 41, 43 classification of, 43 for 55 year old woman, 44 Sharp wave, 41, 43 Sialidosis, 166 Signal processing, 13–14 Silver-chloride electrodes, 17 Simple partial seizure, 180 Simvastatin, 130 Single febrile convulsions, 175 Single generalized convulsion, 172–173 Single seizures, 169–170 Six-and fourteen-Hz-positive bursts, 135 Sleep, 79–94, 125–126. See also Normal EEG of sleep active, 96, 109, 193 deep, 84 ESES, 177 light, 90 paradoxical, 86 POSTS, 82, 126 quiet, 96 slow wave, 84, 177 vertex sharp transients of, 90 Sleep deprivation, 71 Sleep spindles, 82, 92 Slow alpha rhythm, 62 Slow lateral eye movements, 80 Slow posterior walking rhythm, 62 Slow spike-wave, 164 Slow wave sleep, 84, 177 Small, sharp spikes, 133 Somnolence, 254–255 Spasms, 197 Spells, 50–52, 57–58, 61–62, 75–77, 93–94, 111–112, 140–141, 145–146, 148–149, 159–160, 186–187, 224–225 Spikes repetitive/generalized/focal, 264–265 small, sharp, 133 Spinal myoclonus, 272 Spindle coma, 242 Spindle-form morphology, 242 Sporadic delta activity, 54 Squeak phenomenon, 56 SREDA. See Subclinical rhythmic electrographic discharge in adults SSPE. See Subacute sclerosing panencephalitis Staring spells, 50–52, 145–146, 148–149 State-dependence, 48 Static encephalopathy, 271–272

Index

Status epilepticus, 259–276, 263–264 coma after cardiac arrest (74-year-old man), 266–268 coma/myoclonus (45-year-old man), 273–274 coma/myoclonus following cardiac arrest, 275–276 decline in mental status (60-year-old woman), 263–264 epilepsy (52-year-old man), 259–261 static encephalopathy/recurrent myoclonus (8-year-old boy), 271–272 stupor/asterixis (71-year-old man), 269–270 Status myoclonus, 276 Stupor, 232–233, 237–238 Subacute sclerosing panencephalitis (SSPE), 255 Subclinical rhythmic electrographic discharge in adults (SREDA), 183 Subcortical myoclonus, 272 Subdural hematomas, 240 Subdural hygromas, 240 Symmetric driving response, 76–77 Symmetry, 37, 48 Synchrony, 37 hyper, 94 secondary bilateral, 153 Syncopal convulsions, 199 Syncopal myoclonus, 199 Syncope, 199 T Tachypnea, 70 Temporal intermittent rhythmic delta activity (TIRDA), 204 Temporal slowing, 208 Temporal theta bursts, 99 10–20 system, 19–20 Theta coma, 246 Time constant, 7, 8

Index

Timing, 37 Tinnitus, 135–136 TIRDA. See Temporal intermittent rhythmic delta activity Tongue movement, 227 Tonic seizures, 197 Tonic-clonic, 150 Tonic-clonic convulsions, 150–151 Transient, 41 Tremulousness, 182–183 Triphasic waves, 233 True temporal electrodes, 20 Trunk flexion, 159–160 U Unreactive tracings, 100 Urinary tract infection, 53–54 V Vertex sharp transients, 143 Vertex sharp waves, 141 Vertical resolution, 13 Viral meningoencephalitis, 271 Visual hallucinations, 130–131 Visual seizures, 122–124 Voltage, 1, 2 W Wakefulness, 96, 112 Waking. See Normal waking EEG West syndrome, 162 X Xc. See Capacitive reactance Z Zoloft, 63

285